INDEX

Explanations

terms related to government, politics, and authority

oai_token-act-pair · gpt-3.5-turbo

words and phrases indicating grades, ratings, or judgments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 progressing

-0.73

 progression

-0.72

 behaviours

-0.71

 Whilst

-0.69

 analyse

-0.68

 util

-0.68

ilst

-0.68

avour

-0.66

 realise

-0.66

 progresses

-0.65

POSITIVE LOGITS

 Eisenhower

0.80

—

0.80

 Lyndon

0.79

 Nixon

0.77

Newsletter

0.76

—"

0.72

 Americans

0.69

"—

0.69

 seventy

0.68

 Roosevelt

0.68

Activations Density 1.795%

terms related to government, politics, and authority

words and phrases indicating grades, ratings, or judgments

No Comments

No Known Activations