INDEX

Explanations

phrases related to data analysis, statistics, and research studies

oai_token-act-pair · gpt-3.5-turbo

mentions of statistical data and research findings

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

ãĥ¥

-0.59

[/

-0.53

 Shares

-0.50

").

-0.50

obin

-0.47

 crochet

-0.47

 Gors

-0.47

pad

-0.46

ey

-0.46

Ic

-0.46

POSITIVE LOGITS

 meanwhile

0.77

urred

0.67

whose

0.65

 resided

0.64

uilt

0.63

thumbnails

0.59

responsible

0.58

mentioned

0.57

 attest

0.57

 presided

0.57

Activations Density 1.424%

phrases related to data analysis, statistics, and research studies

mentions of statistical data and research findings

No Comments

No Known Activations