INDEX

Explanations

terms that are similar but not identical

oai_token-act-pair · gpt-3.5-turbo

phrases indicating confusion or relationships involving partnerships and associations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 shining

-0.57

nl

-0.57

 stret

-0.57

els

-0.55

agara

-0.54

OUT

-0.54

anda

-0.53

uli

-0.52

 wasted

-0.52

²¾

-0.51

POSITIVE LOGITS

 with

1.21

 WITH

1.15

with

1.04

 With

1.01

 withd

0.97

With

0.95

isons

0.85

nered

0.81

ison

0.81

axter

0.79

Activations Density 0.419%

terms that are similar but not identical

phrases indicating confusion or relationships involving partnerships and associations

No Comments

No Known Activations