INDEX

Explanations

phrases related to the potential impact of actions or events

oai_token-act-pair · gpt-3.5-turbo

phrases related to support and effectiveness in policies or actions

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Elements

-0.72

 towed

-0.64

 nicknamed

-0.63

 handwritten

-0.62

 scanned

-0.62

leted

-0.60

 photographed

-0.59

 gorgeous

-0.59

 spaced

-0.58

 assorted

-0.57

POSITIVE LOGITS

 anymore

1.37

nor

1.18

nor

1.00

 whatsoever

0.98

galitarian

0.95

any

0.89

 unless

0.88

 necessarily

0.88

 anything

0.88

ANY

0.86

Activations Density 0.536%

phrases related to the potential impact of actions or events

phrases related to support and effectiveness in policies or actions

No Comments

No Known Activations