INDEX

Explanations

discussions about controversial actions or policies, particularly those involving ethics and legality

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

terms related to controversial or unethical activities

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

APTER

-0.65

 partName

-0.61

DIT

-0.60

Ô

-0.59

ISSION

-0.59

ãĥª

-0.58

gencies

-0.57

Administ

-0.57

idays

-0.57

Econom

-0.56

POSITIVE LOGITS

 resembling

0.75

 akin

0.70

 allegedly

0.68

 purportedly

0.68

 alongside

0.68

 cheaply

0.66

 supposedly

0.65

 including

0.64

 spiked

0.63

 blamed

0.63

Activations Density 0.730%

discussions about controversial actions or policies, particularly those involving ethics and legality

terms related to controversial or unethical activities

No Comments

No Known Activations