INDEX

Explanations

nouns related to different types of people or characters and their actions or attributes

oai_token-act-pair · gpt-3.5-turbo

phrases suggesting general abilities or behaviors that "anyone" can possess

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

lations

-0.79

 merged

-0.69

 stalled

-0.61

ischer

-0.57

gged

-0.56

 intervals

-0.56

ahime

-0.56

 locality

-0.55

 Sixth

-0.55

izational

-0.55

POSITIVE LOGITS

 imaginable

0.98

 irrespective

0.73

abet

0.71

 capable

0.70

 nowadays

0.70

 except

0.70

 whatsoever

0.69

 regardless

0.69

 shudder

0.66

 alike

0.65

Activations Density 0.224%

nouns related to different types of people or characters and their actions or attributes

phrases suggesting general abilities or behaviors that "anyone" can possess

No Comments

No Known Activations