INDEX

Explanations

years mentioned in various contexts

oai_token-act-pair · gpt-3.5-turbo

specific years and dates

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

ousel

-0.66

elong

-0.65

ardless

-0.65

 genus

-0.64

worldly

-0.62

riet

-0.62

amins

-0.61

onge

-0.60

imated

-0.60

agy

-0.59

POSITIVE LOGITS

 onwards

0.90

 partName

0.81

âĸĪâĸĪ

0.70

 Kass

0.69

 sidx

0.66

GMT

0.65

chev

0.65

 onward

0.64

âĸĪ

0.63

Activations Density 0.080%

years mentioned in various contexts

specific years and dates

No Comments

No Known Activations