INDEX

Explanations

references to notable names, locations, and events

oai_token-act-pair · gpt-3.5-turbo

references to notable public figures, entities, or specific themes

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

 âĢº

-0.69

"]=>

-0.66

ãĥ£

-0.64

cipled

-0.62

Ö

-0.62

rencies

-0.61

orsche

-0.59

Hay

-0.58

Ô

-0.58

hematically

-0.56

POSITIVE LOGITS

 headquarters

0.64

HQ

0.62

 parap

0.59

agine

0.59

 residence

0.59

 tower

0.57

 vicinity

0.57

 Volunteers

0.55

 initiation

0.54

naires

0.54

Activations Density 0.700%

references to notable names, locations, and events

references to notable public figures, entities, or specific themes

No Comments

No Known Activations