INDEX

Explanations

phrases related to events or actions reaching a conclusion or outcome

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate a culmination or outcome

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

esis

-0.66

hee

-0.64

itsch

-0.63

htaking

-0.60

dinand

-0.60

 reclaimed

-0.59

alions

-0.57

 principals

-0.57

esome

-0.57

 Mansion

-0.56

POSITIVE LOGITS

angering

1.13

owment

1.03

ocrin

0.94

angers

0.89

ebted

0.88

orse

0.88

orses

0.86

 abruptly

0.85

ocrine

0.83

 prematurely

0.82

Activations Density 0.045%

phrases related to events or actions reaching a conclusion or outcome

phrases that indicate a culmination or outcome

No Comments

No Known Activations