INDEX

Explanations

keywords related to news and information updates

oai_token-act-pair · gpt-3.5-turbo

references to news and alerts

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

ovember

-0.67

ignt

-0.58

Yin

-0.58

 Laurent

-0.57

 SOLD

-0.56

essee

-0.56

arer

-0.56

 Mortal

-0.56

 grun

-0.56

aughs

-0.55

POSITIVE LOGITS

flash

0.94

room

0.88

worthiness

0.78

worthy

0.73

reader

0.71

 Kavanaugh

0.71

ource

0.69

letters

0.68

bytes

0.68

rooms

0.68

Activations Density 0.025%

keywords related to news and information updates

references to news and alerts

No Comments

No Known Activations