INDEX

Explanations

proper nouns or specific names

oai_token-act-pair · gpt-3.5-turbo

references to past experiences or states of being

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

donald

-0.62

Ãª

-0.62

ire

-0.61

continue

-0.58

apper

-0.55

rone

-0.55

alist

-0.55

ween

-0.53

abouts

-0.53

mpire

-0.52

POSITIVE LOGITS

 however

0.67

 certainly

0.62

 definitely

0.60

 also

0.59

 therefore

0.58

 defin

0.56

 always

0.53

 absolutely

0.51

 obviously

0.50

 very

0.49

Activations Density 0.898%

proper nouns or specific names

references to past experiences or states of being

No Comments

No Known Activations