INDEX

Explanations

phrases related to health risks and medical conditions

oai_token-act-pair · gpt-3.5-turbo

phrases concerning various risks and their 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

ADRA

-0.81

abase

-0.81

REAM

-0.78

Nap

-0.74

estial

-0.74

ulous

-0.73

ele

-0.73

cles

-0.73

ragon

-0.72

ial

-0.69

POSITIVE LOGITS

 exposure

0.84

 contracting

0.82

 harming

0.80

 infection

0.80

 harm

0.79

 respiratory

0.79

 acquiring

0.78

 developing

0.78

 injury

0.78

 litigation

0.78

Activations Density 0.096%

phrases related to health risks and medical conditions

phrases concerning various risks and their implications

No Comments

No Known Activations