INDEX

Explanations

the word "apologize" or its variations.

oai_token-act-pair · gpt-3.5-turbo

terms related to apologies and expressions of remorse

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

weeney

-0.82

arnaev

-0.81

marked

-0.80

picking

-0.74

umbers

-0.71

population

-0.68

aic

-0.67

rooms

-0.66

ulhu

-0.66

markets

-0.66

POSITIVE LOGITS

 unres

1.02

giving

0.93

 apologized

0.93

 apology

0.91

 apologize

0.88

 apologised

0.88

 forgiveness

0.86

 apologizing

0.85

 apologies

0.82

 acknowled

0.81

Activations Density 0.021%

the word "apologize" or its variations.

terms related to apologies and expressions of remorse

No Comments

No Known Activations