INDEX

Explanations

snippets related to statistics and percentages

oai_token-act-pair · gpt-3.5-turbo

quantitative phrases and statistics related to various topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

Reviewer

-0.65

cffff

-0.60

Incre

-0.57

laughs

-0.56

 PROG

-0.56

î

-0.56

 blinding

-0.55

Haunted

-0.55

=>

-0.55

ALLY

-0.53

POSITIVE LOGITS

 involve

1.54

 originate

1.43

 contain

1.25

 occur

1.21

 rely

1.21

 relate

1.17

 fail

1.13

 consist

1.12

 belong

1.11

are

1.10

Activations Density 0.245%

snippets related to statistics and percentages

quantitative phrases and statistics related to various topics

No Comments

No Known Activations