INDEX

Explanations

phrases that provide additional information or context

oai_token-act-pair · gpt-3.5-turbo

the word "although" and its variations, indicating contrast or exception

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

lees

-0.75

enter

-0.75

entry

-0.75

hal

-0.73

oire

-0.72

edu

-0.71

elle

-0.70

unes

-0.70

gem

-0.69

ath

-0.68

POSITIVE LOGITS

 admittedly

0.94

 acknowledging

0.88

 interestingly

0.83

etheless

0.81

 fortunately

0.80

 retaining

0.78

 technically

0.77

chery

0.77

tons

0.76

 preferably

0.75

Activations Density 0.037%

phrases that provide additional information or context

the word "although" and its variations, indicating contrast or exception

No Comments

No Known Activations