INDEX

Explanations

phrases emphasizing completeness or totality

oai_token-act-pair · gpt-3.5-turbo

references to the word "all" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

aminer

-0.69

 Kamp

-0.69

IDS

-0.67

fman

-0.64

SHIP

-0.63

lav

-0.62

nom

-0.60

grad

-0.59

cheon

-0.59

abwe

-0.59

POSITIVE LOGITS

ocating

1.16

igator

1.06

 kinds

1.04

igators

1.02

 sorts

1.02

iances

0.99

owing

0.97

udes

0.96

ocated

0.93

uding

0.93

Activations Density 0.168%

phrases emphasizing completeness or totality

references to the word "all" in various contexts

No Comments

No Known Activations