INDEX

Explanations

judges or judicial-related terms

oai_token-act-pair · gpt-3.5-turbo

mentions of judges and their rulings

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

ulus

-0.76

otin

-0.72

orne

-0.71

rolet

-0.70

iliar

-0.69

ulates

-0.67

ulated

-0.67

urst

-0.66

 IMAGES

-0.66

haar

-0.64

POSITIVE LOGITS

 Advocate

0.84

 rulings

0.81

 transcripts

0.77

 presiding

0.76

rooms

0.76

 Judy

0.75

 Gorsuch

0.75

redd

0.74

Judge

0.71

 overseeing

0.69

Activations Density 0.016%

judges or judicial-related terms

mentions of judges and their rulings

No Comments

No Known Activations