INDEX

Explanations

mentions of specific actions or situations

oai_token-act-pair · gpt-3.5-turbo

instances of listening or hearing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

never

-0.70

;;;;;;;;;;;;

-0.70

;;;;;;;;

-0.69

 nonetheless

-0.69

 redes

-0.65

although

-0.64

 complied

-0.64

etheless

-0.63

 remained

-0.62

 nevertheless

-0.62

POSITIVE LOGITS

 somebody

0.72

 isEnabled

0.71

 someone

0.69

pires

0.60

someone

0.58

 adversity

0.56

 something

0.56

 disrespect

0.55

context

0.54

new

0.54

Activations Density 0.352%

mentions of specific actions or situations

instances of listening or hearing

No Comments

No Known Activations