INDEX

Explanations

the word "only" with a high level of activation

oai_token-act-pair · gpt-3.5-turbo

the word "only" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

insula

-0.89

idon

-0.76

finder

-0.67

 heterogeneity

-0.66

Nev

-0.63

Dru

-0.62

multipl

-0.60

 Dynamics

-0.56

 Intern

-0.56

 fragmentation

-0.54

POSITIVE LOGITS

 marginally

0.96

 incidentally

0.69

 scratched

0.69

phies

0.69

 kidding

0.68

 mildly

0.67

ices

0.67

 scratches

0.66

isons

0.65

 limited

0.65

Activations Density 0.062%

the word "only" with a high level of activation

the word "only" and its variations

No Comments

No Known Activations