INDEX

Explanations

words related to being attracted or pulled towards something

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "drawn" in various contexts

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

icans

-0.73

ican

-0.73

 Canad

-0.72

 Stability

-0.69

ership

-0.65

FK

-0.64

ica

-0.62

cies

-0.62

gram

-0.60

ernel

-0.60

POSITIVE LOGITS

 irresist

0.91

 drawn

0.83

 toward

0.82

 towards

0.81

 into

0.77

 aback

0.76

 Drawn

0.75

entious

0.75

 Variant

0.73

 Into

0.73

Activations Density 0.032%

words related to being attracted or pulled towards something

occurrences of the word "drawn" in various contexts

No Comments

No Known Activations