INDEX

Explanations

references to the term "Wolf" at different levels of specificity, possibly related to a specific topic or entity named "Wolf"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Wolf" in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

the word "Wolf" regardless of context

oai_token-act-pair · gpt-4-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

bably

-0.91

pora

-0.81

acterial

-0.80

ntil

-0.77

acists

-0.77

thritis

-0.73

ursed

-0.73

 tradem

-0.73

senal

-0.72

llular

-0.72

POSITIVE LOGITS

 Wolf

1.29

enstein

1.26

hound

1.13

Wolf

1.13

sburg

1.06

 Wolves

1.03

pack

0.98

owitz

0.98

 Fenrir

0.95

bats

0.94

Activations Density 0.007%

references to the term "Wolf" at different levels of specificity, possibly related to a specific topic or entity named "Wolf"

occurrences of the word "Wolf" in various contexts

the word "Wolf" regardless of context

No Comments

No Known Activations