INDEX

Explanations

numbers and technical terms

oai_token-act-pair · gpt-3.5-turbo

a mix of numeric values, special characters, and mentions of individuals or locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

DragonMagazine

-0.68

 glim

-0.64

 resil

-0.64

 hindsight

-0.60

leck

-0.58

 leaps

-0.57

é¾įåĸļå£«

-0.57

CRIPTION

-0.57

reau

-0.56

pedia

-0.56

POSITIVE LOGITS

 thirds

0.62

 choke

0.55

ociate

0.55

thirds

0.55

ukemia

0.55

Osc

0.53

cheat

0.52

 Seym

0.52

ovych

0.51

 husband

0.50

Activations Density 1.358%

numbers and technical terms

a mix of numeric values, special characters, and mentions of individuals or locations

No Comments

No Known Activations