INDEX

Explanations

expressions related to assessment and evaluation

oai_token-act-pair · gpt-3.5-turbo

discussions of legal consequences and policy implications

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

 Chili

-0.68

bara

-0.63

da

-0.62

 Trailer

-0.61

 Bord

-0.61

Soc

-0.61

 Vive

-0.61

 Fortune

-0.59

 este

-0.59

mosp

-0.58

POSITIVE LOGITS

 themselves

0.98

 collectively

0.93

 individually

0.91

ensitive

0.73

 systematically

0.73

geries

0.71

 jointly

0.70

 constitute

0.70

 include

0.69

 originate

0.69

Activations Density 0.962%

expressions related to assessment and evaluation

discussions of legal consequences and policy implications

No Comments

No Known Activations