INDEX

Explanations

phrases related to official reports and statements

oai_token-act-pair · gpt-3.5-turbo

mention of reports or official findings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

à¹

-0.74

 Himself

-0.73

ãĥŀ

-0.72

 Ghostbusters

-0.71

IME

-0.68

alogue

-0.64

icio

-0.64

ãĥ¬

-0.62

 souven

-0.62

agram

-0.61

POSITIVE LOGITS

Recommend

0.92

 findings

0.85

 methodological

0.83

 systematic

0.81

 respondents

0.81

 gaps

0.76

ificantly

0.76

 statistically

0.75

 disparities

0.75

 citing

0.72

Activations Density 0.440%

phrases related to official reports and statements

mention of reports or official findings

No Comments

No Known Activations