INDEX

Explanations

actions of transforming or converting things into something else

oai_token-act-pair · gpt-3.5-turbo

phrases about transformation or conversion processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

cation

-0.77

inately

-0.73

ritz

-0.72

erity

-0.70

ran

-0.69

ername

-0.68

enance

-0.66

antage

-0.65

raint

-0.64

no

-0.64

POSITIVE LOGITS

 usable

0.88

 something

0.73

 profitable

0.65

ãĥ¼ãĥ

0.64

quished

0.64

AFTA

0.62

0.61

Obj

0.61

 surrogate

0.59

 productive

0.59

Activations Density 0.071%

actions of transforming or converting things into something else

phrases about transformation or conversion processes

No Comments

No Known Activations