INDEX

Explanations

queries or instructions related to technology or procedures

oai_token-act-pair · gpt-3.5-turbo

questions and inquiries about various topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 rhet

-0.71

Putin

-0.70

 damned

-0.67

thought

-0.65

ously

-0.65

pants

-0.65

gypt

-0.63

illions

-0.63

."[

-0.62

sein

-0.61

POSITIVE LOGITS

?:

1.10

Use

1.01

 Does

0.95

 Particip

0.92

 Include

0.92

 Apply

0.92

 Uses

0.91

 Should

0.91

 Choose

0.89

 Supported

0.89

Activations Density 0.184%

queries or instructions related to technology or procedures

questions and inquiries about various topics

No Comments

No Known Activations