INDEX

Explanations

phrases relating to locations and distances

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate approximate quantities or timings

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hesis

-0.76

oven

-0.72

chens

-0.72

BT

-0.67

ammy

-0.65

PRESS

-0.64

ãĤ¿

-0.64

ÂłÂłÂłÂłÂłÂłÂłÂł

-0.63

ÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂł

-0.63

gallery

-0.63

POSITIVE LOGITS

 halfway

0.80

aleb

0.70

eus

0.70

 midway

0.67

 pregn

0.67

 Noon

0.65

 calib

0.64

 respecting

0.64

 migrating

0.62

how

0.61

Activations Density 0.121%

phrases relating to locations and distances

phrases that indicate approximate quantities or timings

No Comments

No Known Activations