INDEX

Explanations

terms related to display configurations or sizes

oai_token-act-pair · gpt-3.5-turbo

terms related to visual display configurations and dimensions

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

Ń·

-0.75

ubi

-0.64

 Lime

-0.63

 Chel

-0.62

 mete

-0.62

imaru

-0.61

 Whitman

-0.60

ÅĤ

-0.58

 perman

-0.58

apest

-0.58

POSITIVE LOGITS

empt

0.83

fledged

0.77

heartedly

0.72

insula

0.70

pection

0.69

Initialized

0.69

(>

0.68

ledged

0.66

cise

0.66

tions

0.66

Activations Density 0.068%

terms related to display configurations or sizes

terms related to visual display configurations and dimensions

No Comments

No Known Activations