INDEX

Explanations

numerical values, such as distances, quantities, and measurements

oai_token-act-pair · gpt-3.5-turbo

quantitative data and statistics

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

chwitz

-0.67

tainment

-0.67

£ı

-0.67

 Emblem

-0.65

ĵĺ

-0.63

é¾įå¥ĳå£«

-0.62

ħĭ

-0.62

ģĸ

-0.62

 Muse

-0.61

Agent

-0.60

POSITIVE LOGITS

cm

0.88

 minutes

0.83

Â½

0.80

 nights

0.75

0.73

 days

0.72

 strokes

0.72

 decimal

0.71

leigh

0.71

mins

0.71

Activations Density 0.289%

numerical values, such as distances, quantities, and measurements

quantitative data and statistics

No Comments

No Known Activations