INDEX

Explanations

phrases related to specific names or identifiers

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

advertisement

-0.65

ngth

-0.63

ritch

-0.62

riks

-0.58

apache

-0.57

 Artifact

-0.56

Reviewer

-0.56

URA

-0.56

ascus

-0.56

 benign

-0.55

POSITIVE LOGITS

aroo

0.90

zona

0.69

dinand

0.68

hani

0.68

owitz

0.67

Ã£

0.66

oola

0.65

 Alto

0.64

oche

0.64

Topic

0.64

Activations Density 0.084%

phrases related to specific names or identifiers

No Comments

No Known Activations