INDEX

Explanations

references to specific years or decades

oai_token-act-pair · gpt-3.5-turbo

references to specific decades and years

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 bearer

-0.65

 recip

-0.58

 Quote

-0.57

PLA

-0.56

 Dinosaur

-0.56

 tribunal

-0.56

 orig

-0.55

aukee

-0.54

NCT

-0.54

ingen

-0.54

POSITIVE LOGITS

ixties

1.04

ties

0.94

hower

0.86

 twenties

0.82

rils

0.80

ugh

0.80

liction

0.78

rence

0.77

#$

0.75

cture

0.74

Activations Density 0.106%

references to specific years or decades

references to specific decades and years

No Comments

No Known Activations