INDEX

Explanations

references to the term "Wolf"

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "Wolf" and its variations, indicating a focus on that specific subject or character

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

bably

-0.72

nces

-0.71

thritis

-0.70

acists

-0.70

 seiz

-0.70

 fuse

-0.67

encies

-0.66

 contrace

-0.66

quo

-0.62

ITY

-0.61

POSITIVE LOGITS

enstein

1.36

sburg

1.33

hound

1.27

gang

1.07

pack

1.07

ram

1.06

owitz

1.03

endor

1.01

rider

0.95

bats

0.93

Activations Density 0.036%

references to the term "Wolf"

mentions of the word "Wolf" and its variations, indicating a focus on that specific subject or character

No Comments

No Known Activations