INDEX

Explanations

instances where personal information or actions are being discussed or reported

oai_token-act-pair · gpt-3.5-turbo

references to provided information or documentation

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

 Metatron

-0.66

 Quentin

-0.63

readable

-0.60

tein

-0.58

 stal

-0.57

cop

-0.56

 Adin

-0.56

IB

-0.56

 Indust

-0.56

ocide

-0.55

POSITIVE LOGITS

abouts

0.72

uggle

0.67

pez

0.66

amines

0.66

ults

0.62

ioned

0.62

ETS

0.60

itud

0.60

ago

0.60

ãĥ¼ãĤ¯

0.60

Activations Density 0.416%

instances where personal information or actions are being discussed or reported

references to provided information or documentation

No Comments

No Known Activations