INDEX

Explanations

sequential events or actions

oai_token-act-pair · gpt-3.5-turbo

phrases related to work schedules and shifts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

nor

-0.68

wrong

-0.61

Yet

-0.61

PIN

-0.60

BuyableInstoreAndOnline

-0.60

than

-0.59

 Didn

-0.58

certain

-0.57

"]=>

-0.56

 lately

-0.54

POSITIVE LOGITS

 thereafter

1.13

 culminating

1.06

 afterwards

0.99

 whence

0.97

 later

0.93

 followed

0.93

 onwards

0.86

 where

0.83

 afterward

0.82

 wherein

0.81

Activations Density 0.684%

sequential events or actions

phrases related to work schedules and shifts

No Comments

No Known Activations