INDEX

Explanations

ambiguous or unspecified nouns related to actions or scenarios

oai_token-act-pair · gpt-3.5-turbo

phrases expressing uncertainty or vagueness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ories

-0.67

conn

-0.66

icons

-0.65

adapt

-0.63

 Colomb

-0.62

iPhone

-0.60

istors

-0.60

kens

-0.60

 fathers

-0.59

 Lithuania

-0.58

POSITIVE LOGITS

 else

0.97

 thereof

0.94

Else

0.90

¥µ

0.89

etc

0.85

ĪĴ

0.84

 similar

0.80

whatever

0.80

Ń

0.78

etheless

0.77

Activations Density 0.102%

ambiguous or unspecified nouns related to actions or scenarios

phrases expressing uncertainty or vagueness

No Comments

No Known Activations