INDEX

Explanations

words related to regulations or requirements

oai_token-act-pair · gpt-3.5-turbo

references to eligibility in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

EMS

-0.74

âĸ¬

-0.72

;;;;;;;;

-0.67

âķĲâķĲ

-0.65

âĸ¬âĸ¬

-0.65

 beating

-0.64

VIDEO

-0.64

Build

-0.63

SO

-0.63

Merc

-0.62

POSITIVE LOGITS

 eligibility

1.28

 criteria

0.97

 qualifications

0.89

 clearance

0.88

 ineligible

0.87

 disqualified

0.86

 eligible

0.85

 credentials

0.82

 requirements

0.79

ibility

0.79

Activations Density 0.006%

words related to regulations or requirements

references to eligibility in various contexts

No Comments

No Known Activations