INDEX

Explanations

statistical data and facts

oai_token-act-pair · gpt-3.5-turbo

numeric data and statistics related to polling and survey results

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 prosec

-0.93

 describ

-0.92

 tiss

-0.86

 barr

-0.86

 symp

-0.83

 indo

-0.83

 boun

-0.82

 cour

-0.82

 mutually

-0.80

 dynam

-0.79

POSITIVE LOGITS

Overall

1.95

Average

1.85

Among

1.82

Percent

1.80

Across

1.80

Compared

1.79

Nearly

1.78

Between

1.74

Of

1.73

Researchers

1.68

Activations Density 0.334%

statistical data and facts

numeric data and statistics related to polling and survey results

No Comments

No Known Activations