INDEX

Explanations

time-related indications, specifically the term "months."

oai_token-act-pair · gpt-3.5-turbo

references to the duration of time, particularly in months and years

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

agonist

-0.66

isin

-0.63

Exit

-0.63

ocate

-0.62

 sphere

-0.61

Indiana

-0.59

ICO

-0.59

 exception

-0.59

Jackson

-0.59

ocard

-0.58

POSITIVE LOGITS

ago

0.93

entimes

0.86

zens

0.79

 overdue

0.79

 millennia

0.74

ankind

0.71

 overlooked

0.70

uly

0.69

coni

0.68

ohn

0.65

Activations Density 0.059%

time-related indications, specifically the term "months."

references to the duration of time, particularly in months and years

No Comments

No Known Activations