INDEX

Explanations

expressions related to time

oai_token-act-pair · gpt-3.5-turbo

references to specific time periods or repeated mentions of time

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

avorite

-0.71

 cipher

-0.70

 plunge

-0.67

KING

-0.67

ortium

-0.66

malink

-0.66

mology

-0.64

 WORLD

-0.63

 mathemat

-0.63

MODE

-0.61

POSITIVE LOGITS

frames

0.85

glass

0.72

opes

0.67

oshenko

0.67

staff

0.66

cross

0.64

iday

0.64

frame

0.62

 distortions

0.62

oa

0.61

Activations Density 0.019%

expressions related to time

references to specific time periods or repeated mentions of time

No Comments

No Known Activations