INDEX

Explanations

words related to specific events or occurrences

oai_token-act-pair · gpt-3.5-turbo

transitional phrases and connectors that indicate a sequence or contrast in information

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

arthy

-0.61

],"

-0.60

existent

-0.59

"}],"

-0.58

emale

-0.58

farious

-0.57

',"

-0.56

querade

-0.56

è¦ļéĨĴ

-0.53

foundland

-0.53

POSITIVE LOGITS

cknowled

0.71

cing

0.54

oret

0.46

 Leader

0.46

Sync

0.46

 Finder

0.45

 Spawn

0.45

 Reck

0.44

 Issue

0.44

 Enemy

0.44

Activations Density 0.563%

words related to specific events or occurrences

transitional phrases and connectors that indicate a sequence or contrast in information

No Comments

No Known Activations