INDEX

Explanations

revisions or changes in documents or plans

oai_token-act-pair · gpt-3.5-turbo

references to revisions or updates in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ships

-0.89

rage

-0.79

culture

-0.78

OPA

-0.77

aceae

-0.74

hold

-0.73

bara

-0.73

bearing

-0.72

abet

-0.72

acht

-0.71

POSITIVE LOGITS

 versions

0.93

escription

0.90

ieval

0.86

iterranean

0.78

 editions

0.74

ouble

0.70

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

0.69

efined

0.68

iated

0.68

nesday

0.67

Activations Density 0.070%

revisions or changes in documents or plans

references to revisions or updates in various contexts

No Comments

No Known Activations