INDEX

Explanations

phrases indicating the location or discovery of something

oai_token-act-pair · gpt-3.5-turbo

instances of the preposition "in" indicating location or context

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

%%

-0.64

 pals

-0.61

—-

-0.61

 PLEASE

-0.61

 instr

-0.61

LOG

-0.61

ens

-0.60

emo

-0.60

username

-0.60

ctor

-0.60

POSITIVE LOGITS

 conjunction

1.23

ordinate

1.21

 accordance

1.16

organic

1.15

humane

1.12

 lieu

1.12

animate

1.11

clusions

1.11

escap

1.05

roads

0.99

Activations Density 0.267%

phrases indicating the location or discovery of something

instances of the preposition "in" indicating location or context

No Comments

No Known Activations