INDEX

Explanations

numbers at the end of words or phrases

oai_token-act-pair · gpt-3.5-turbo

the character 'ľ' in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 lapt

-0.77

 disadvant

-0.76

 scissors

-0.75

 levers

-0.75

 condem

-0.73

matic

-0.72

 machines

-0.72

 hemor

-0.71

raints

-0.71

 sails

-0.70

POSITIVE LOGITS

âĶĢâĶĢ

1.32

ï¸ı

1.09

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

1.07

âĶĢâĶĢâĶĢâĶĢ

0.99

conom

0.89

Â°

0.84

×

0.82

0.81

×Ķ

0.79

âĸł

0.79

Activations Density 0.138%

numbers at the end of words or phrases

the character 'ľ' in the text

No Comments

No Known Activations