INDEX

Explanations

terms related to the word 'ets' followed by a number

oai_token-act-pair · gpt-3.5-turbo

references to various types of "ets."

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 exha

-0.80

 livest

-0.72

shield

-0.71

 surv

-0.70

 unden

-0.70

âĸ¬

-0.68

 gobl

-0.66

 unification

-0.65

shire

-0.65

 undermin

-0.64

POSITIVE LOGITS

wana

1.15

pace

1.12

ail

1.06

cape

1.04

ilon

1.04

heet

1.01

tered

0.96

onic

0.95

ets

0.95

ails

0.95

Activations Density 0.016%

terms related to the word 'ets' followed by a number

references to various types of "ets."

No Comments

No Known Activations