INDEX

Explanations

lists of items

oai_token-act-pair · gpt-3.5-turbo

the word "list" 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 @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

selves

-0.72

 whist

-0.65

 Aber

-0.64

 rhyth

-0.63

 leisure

-0.60

 Trade

-0.60

 Train

-0.59

 hearts

-0.59

vertisement

-0.58

 recess

-0.58

POSITIVE LOGITS

erv

1.04

lists

0.81

witz

0.80

 listings

0.79

 listing

0.78

erve

0.77

list

0.77

 lists

0.75

uably

0.75

 criteria

0.75

Activations Density 0.027%

lists of items

the word "list" in various contexts

No Comments

No Known Activations