INDEX

Explanations

instances where the concept of "fairness" is mentioned

oai_token-act-pair · gpt-3.5-turbo

instances of the concept of fairness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

CHAT

-0.83

apse

-0.81

uality

-0.72

 Saga

-0.66

 Reincarnated

-0.66

Ultra

-0.65

clips

-0.65

\<

-0.65

OUS

-0.64

Extra

-0.62

POSITIVE LOGITS

grounds

1.25

yt

1.10

 fair

0.90

ground

0.90

fair

0.89

iciary

0.83

abouts

0.81

itably

0.79

child

0.79

heet

0.77

Activations Density 0.015%

instances where the concept of "fairness" is mentioned

instances of the concept of fairness

No Comments

No Known Activations