INDEX

Explanations

references to specific entities or names, potentially related to lawsuits or legal proceedings

oai_token-act-pair · gpt-3.5-turbo

instances of the letter 'L' in various contexts

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

 decomp

-0.82

 JPEG

-0.73

 contempor

-0.64

 liter

-0.64

 Gmail

-0.62

 Mobil

-0.61

©¶æ

-0.60

 scrut

-0.59

 itiner

-0.58

rag

-0.58

POSITIVE LOGITS

1.37

tal

1.07

ship

1.07

ses

1.03

swer

1.02

ski

0.94

tarian

0.94

ships

0.91

ties

0.90

ty

0.89

Activations Density 0.272%

references to specific entities or names, potentially related to lawsuits or legal proceedings

instances of the letter 'L' in various contexts

No Comments

No Known Activations