INDEX

Explanations

phrases related to specific actions or tasks that often involve technical or procedural instructions

oai_token-act-pair · gpt-3.5-turbo

the end of a document or section break

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Seym

-0.67

 vulner

-0.65

 Niet

-0.64

tun

-0.63

 disadvant

-0.61

 prest

-0.60

calling

-0.59

 secretaries

-0.59

paren

-0.59

anamo

-0.58

POSITIVE LOGITS

 RELEASE

0.83

 celebrate

0.83

pload

0.79

ilet

0.76

 provide

0.75

asted

0.74

asters

0.73

 capitalize

0.72

ggles

0.72

 accelerate

0.72

Activations Density 0.039%

phrases related to specific actions or tasks that often involve technical or procedural instructions

the end of a document or section break

No Comments

No Known Activations