INDEX

Explanations

verbs in the gerund form (ending in '-ing') indicating ongoing actions or activities

oai_token-act-pair · gpt-3.5-turbo

instances of the word "doing" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Tier

-0.81

Tai

-0.73

rum

-0.69

lights

-0.69

tight

-0.68

oufl

-0.67

 Taiwan

-0.67

 Handling

-0.65

 Indonesia

-0.64

 Leth

-0.64

POSITIVE LOGITS

 mosqu

0.88

pez

0.86

VIDIA

0.82

 toget

0.81

omething

0.81

oley

0.79

issance

0.77

ezvous

0.73

 engagements

0.71

 transact

0.71

Activations Density 0.051%

verbs in the gerund form (ending in '-ing') indicating ongoing actions or activities

instances of the word "doing" in various contexts

No Comments

No Known Activations