INDEX

Explanations

phrases related to policies, decisions, and procedures

oai_token-act-pair · gpt-3.5-turbo

instances of significant events or themes

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

$.

-0.69

 gobl

-0.63

 Niet

-0.62

 hemor

-0.61

 mathemat

-0.61

SPONSORED

-0.58

+.

-0.58

 cannabin

-0.57

utm

-0.56

".[

-0.56

POSITIVE LOGITS

 ensures

0.72

 helps

0.72

 requires

0.70

Â¶

0.68

 yourself

0.68

 improves

0.67

 Yourself

0.67

 implies

0.67

 entails

0.66

rift

0.65

Activations Density 0.745%

phrases related to policies, decisions, and procedures

instances of significant events or themes

No Comments

No Known Activations