INDEX

Explanations

specific character sequences, potentially related to titles or names

oai_token-act-pair · gpt-3.5-turbo

the character 'Ļ' 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 @ 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

 disadvant

-0.95

 contrace

-0.94

 mathemat

-0.93

 Palestin

-0.87

 misunder

-0.81

 vulner

-0.80

 smugglers

-0.80

 pestic

-0.79

 fortun

-0.79

 satell

-0.78

POSITIVE LOGITS

ï¸ı

1.19

ï¸

0.99

tre

0.97

Hol

0.93

ski

0.87

ship

0.84

İ

0.80

lime

0.80

lu

0.77

VICE

0.77

Activations Density 0.309%

specific character sequences, potentially related to titles or names

the character 'Ļ' in various contexts

No Comments

No Known Activations