INDEX

Explanations

statistics or quantities of events happening annually

oai_token-act-pair · gpt-3.5-turbo

references to occurrences and statistics that happen annually

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

anooga

-0.73

ymes

-0.69

anamo

-0.69

urrection

-0.67

gotten

-0.66

ells

-0.66

aleb

-0.65

ect

-0.65

reprene

-0.65

merce

-0.64

POSITIVE LOGITS

 thereafter

0.93

ago

0.81

long

0.81

lins

0.74

Iw

0.71

 basis

0.70

lot

0.69

summary

0.68

Ago

0.68

Cox

0.67

Activations Density 0.035%

statistics or quantities of events happening annually

references to occurrences and statistics that happen annually

No Comments

No Known Activations