INDEX

Explanations

positive adjectives or expressions of admiration

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "great."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ople

-0.73

ilus

-0.72

eter

-0.69

bus

-0.67

SPONSORED

-0.66

pex

-0.66

Downloadha

-0.64

mitt

-0.64

former

-0.63

cling

-0.63

POSITIVE LOGITS

 strides

0.89

sword

0.88

 opportunity

0.85

 outdoors

0.83

 deal

0.81

 tasting

0.79

 appreciation

0.77

 Dane

0.75

 accomplishment

0.75

 introductory

0.74

Activations Density 0.056%

positive adjectives or expressions of admiration

occurrences of the word "great."

No Comments

No Known Activations