INDEX

Explanations

references to legal professions, specifically coroners and examiners

oai_token-act-pair · gpt-3.5-turbo

terms related to medical examination and investigation officials

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

ern

-0.76

este

-0.74

 Bengal

-0.73

eco

-0.72

amy

-0.69

shaw

-0.69

Ka

-0.68

Kap

-0.68

wal

-0.68

POSITIVE LOGITS

 rall

0.84

 confir

0.80

oner

0.78

 psychiat

0.77

ipolar

0.76

 coroner

0.76

 isot

0.76

 suspic

0.75

 examiner

0.71

 occas

0.71

Activations Density 0.036%

references to legal professions, specifically coroners and examiners

terms related to medical examination and investigation officials

No Comments

No Known Activations