INDEX

Explanations

references to the ability or permission to do something that enables progress

oai_token-act-pair · gpt-3.5-turbo

pronouns indicating people or entities involved in actions or situations

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

fect

-0.88

 Siege

-0.74

town

-0.73

 Speedway

-0.69

 Onion

-0.69

Atlantic

-0.68

vertisements

-0.65

fleet

-0.64

 Words

-0.64

 Carlton

-0.64

POSITIVE LOGITS

 personally

0.90

 undertake

0.76

own

0.75

 retain

0.74

selves

0.73

 preferred

0.73

 perceive

0.72

 invested

0.71

 tremend

0.71

 feel

0.71

Activations Density 0.196%

references to the ability or permission to do something that enables progress

pronouns indicating people or entities involved in actions or situations

No Comments

No Known Activations