INDEX

Explanations

words related to long-term impacts or consequences

oai_token-act-pair · gpt-3.5-turbo

references to long-term versus short-term effects

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

 CODE

-0.66

 VIDEOS

-0.65

othing

-0.65

Cod

-0.64

ymes

-0.64

ideshow

-0.61

MET

-0.61

leck

-0.60

icken

-0.59

MSN

-0.59

POSITIVE LOGITS

term

1.04

 term

1.00

 haul

0.92

 afterlife

0.84

 aftermath

0.83

 confines

0.81

 absence

0.80

 future

0.79

est

0.78

 meantime

0.78

Activations Density 0.049%

words related to long-term impacts or consequences

references to long-term versus short-term effects

No Comments

No Known Activations