INDEX

Explanations

negations and actions related to decision-making

oai_token-act-pair · gpt-3.5-turbo

negative contractions and their usage in sentences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

-0.75

Solution

-0.74

forcement

-0.73

 nonetheless

-0.73

arb

-0.72

POL

-0.70

etimes

-0.67

onomy

-0.67

ãĤ¼

-0.67

evidence

-0.66

POSITIVE LOGITS

 typo

0.93

 boring

0.92

shy

0.91

 spoilers

0.90

 fancy

0.89

 duplication

0.88

 bulky

0.87

 gimm

0.85

 pricey

0.83

 disappoint

0.81

Activations Density 0.575%

negations and actions related to decision-making

negative contractions and their usage in sentences

No Comments

No Known Activations