INDEX

Explanations

ALBERTA's token identifiers are still activated

oai_token-act-pair · gpt-3.5-turbo

the symbol "Ŀ" used in various contexts throughout the document

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

-0.70

 obser

-0.69

 resemb

-0.66

 ende

-0.66

 resemblance

-0.65

 controvers

-0.65

 imagination

-0.64

 likeness

-0.64

 contrace

-0.64

izen

-0.63

POSITIVE LOGITS

ï¸ı

1.03

Â¯

0.99

ï¸

0.87

âĢł

0.80

Â°

0.80

ÃĽ

0.77

âĸ

0.77

tra

0.74

âģ

0.74

âĻ¦

0.74

Activations Density 0.188%

ALBERTA's token identifiers are still activated

the symbol "Ŀ" used in various contexts throughout the document

No Comments

No Known Activations