INDEX

Explanations

words related to negative behavior or conditions

oai_token-act-pair · gpt-3.5-turbo

terminology related to negative actions or consequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 carrot

-0.79

 orally

-0.74

 readable

-0.74

 bartender

-0.71

 goat

-0.70

cyn

-0.67

 foremost

-0.67

 boxed

-0.66

 translator

-0.66

 interpreter

-0.66

POSITIVE LOGITS

ancies

1.52

ations

1.50

encies

1.47

rences

1.41

ances

1.41

acies

1.36

ings

1.32

uations

1.32

inations

1.29

abilities

1.28

Activations Density 0.227%

words related to negative behavior or conditions

terminology related to negative actions or consequences

No Comments

No Known Activations