INDEX

Explanations

references to full-length or comprehensive content, such as movies, texts, or productions

oai_token-act-pair · gpt-3.5-turbo

references to complete or "full" experiences or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Lowe

-0.64

 misplaced

-0.62

Phi

-0.61

 pige

-0.60

bably

-0.60

 conspicuous

-0.59

 Lilly

-0.59

 widely

-0.58

 Plat

-0.58

gex

-0.57

POSITIVE LOGITS

ledged

0.89

fledged

0.75

 Coverage

0.73

iability

0.71

 Jacket

0.68

(>

0.66

 coverage

0.66

illon

0.66

load

0.64

 spectrum

0.63

Activations Density 0.227%

references to full-length or comprehensive content, such as movies, texts, or productions

references to complete or "full" experiences or entities

No Comments

No Known Activations