INDEX

Explanations

words related to changes or modifications

oai_token-act-pair · gpt-3.5-turbo

terms related to changes or modifications in state or condition

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

hang

-0.86

abet

-0.84

yer

-0.78

lay

-0.77

ships

-0.75

went

-0.74

where

-0.74

abilia

-0.74

goal

-0.73

MAT

-0.73

POSITIVE LOGITS

 versions

0.95

 subur

0.91

iterranean

0.88

ieval

0.86

 aback

0.81

 satell

0.79

iated

0.78

 hijacked

0.77

 pione

0.76

nesday

0.75

Activations Density 0.078%

words related to changes or modifications

terms related to changes or modifications in state or condition

No Comments

No Known Activations