INDEX

Explanations

words related to "ace."

oai_token-act-pair · gpt-3.5-turbo

the word "ace" 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 @ 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

ostic

-0.72

icum

-0.66

 clinically

-0.64

amins

-0.63

aging

-0.61

agers

-0.61

 FANT

-0.61

 medically

-0.60

 taxing

-0.60

ect

-0.60

POSITIVE LOGITS

llular

0.99

llor

0.92

lyn

0.87

lla

0.82

idon

0.82

ments

0.80

flight

0.80

llan

0.76

lect

0.74

lets

0.73

Activations Density 0.027%

words related to "ace."

the word "ace" in various contexts

No Comments

No Known Activations