INDEX

Explanations

terminology related to statistics or numerical values within text

oai_token-act-pair · gpt-3.5-turbo

recurring themes and numerical representations in political discussions

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

 scrap

-0.61

 equivalent

-0.60

 wedd

-0.60

 frontline

-0.60

 covenant

-0.59

 equival

-0.59

ĸļ

-0.59

 herself

-0.59

 incent

-0.59

 amph

-0.59

POSITIVE LOGITS

ccording

1.09

If

0.99

Contribut

0.98

Advertisement

0.98

Topics

0.98

Official

0.98

There

0.97

Since

0.97

Whether

0.96

However

0.96

Activations Density 0.907%

terminology related to statistics or numerical values within text

recurring themes and numerical representations in political discussions

No Comments

No Known Activations