INDEX

Explanations

terms related to organizing or arranging things in a specific order

oai_token-act-pair · gpt-3.5-turbo

terms related to sorting and organization processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

psey

-0.87

stood

-0.80

interrupted

-0.75

roach

-0.75

PLIED

-0.73

pires

-0.73

ewitness

-0.71

intosh

-0.70

LAN

-0.69

runner

-0.69

POSITIVE LOGITS

 sorting

1.03

ãĤ¦ãĤ¹

0.88

 Parenthood

0.85

ãĥĥ

0.84

 partName

0.80

Īè

0.80

Sort

0.79

 Sort

0.77

 alphabet

0.77

 sorted

0.71

Activations Density 0.015%

terms related to organizing or arranging things in a specific order

terms related to sorting and organization processes

No Comments

No Known Activations