INDEX

Explanations

proper nouns, specifically related to the legal field including judges, legal cases, and legal institutions

oai_token-act-pair · gpt-3.5-turbo

terms related to judgment and legal concepts

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

rison

-0.81

ocene

-0.81

assing

-0.80

asses

-0.78

assed

-0.76

atha

-0.75

 Alto

-0.75

asso

-0.75

azine

-0.73

 stadiums

-0.73

POSITIVE LOGITS

è¯

0.85

ividual

0.77

CLASSIFIED

0.73

IVERS

0.70

ktop

0.68

nik

0.68

ensional

0.68

yers

0.68

soever

0.67

ISION

0.66

Activations Density 0.082%

proper nouns, specifically related to the legal field including judges, legal cases, and legal institutions

terms related to judgment and legal concepts

No Comments

No Known Activations