INDEX

Explanations

words related to changes, reforms, and updates

oai_token-act-pair · gpt-3.5-turbo

references to the concept of change or modifications

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

english

-0.75

vey

-0.70

gdala

-0.67

utherland

-0.66

WOOD

-0.65

acebook

-0.64

FIELD

-0.64

umbo

-0.64

horn

-0.63

bon

-0.63

POSITIVE LOGITS

 drastic

0.88

 gradual

0.84

redo

0.84

 incremental

0.80

 delet

0.80

 restructuring

0.80

 additions

0.79

 rewriting

0.79

 undone

0.79

 enhancements

0.78

Activations Density 0.586%

words related to changes, reforms, and updates

references to the concept of change or modifications

No Comments

No Known Activations