INDEX

Explanations

statistical or numerical information

oai_token-act-pair · gpt-3.5-turbo

indicators of public policy and governance

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Supports

-0.56

 IMAGES

-0.56

 Tours

-0.52

 PLEASE

-0.50

 Talks

-0.49

inav

-0.48

ossom

-0.48

 Whilst

-0.47

 Telescope

-0.47

 Calls

-0.47

POSITIVE LOGITS

 indistinguishable

0.75

 merely

0.72

 considerably

0.72

 relegated

0.71

 virtually

0.70

 downright

0.70

 deemed

0.67

 considered

0.66

Ļ½

0.64

 exceedingly

0.64

Activations Density 0.760%

statistical or numerical information

indicators of public policy and governance

No Comments

No Known Activations