INDEX

Explanations

phrases related to instructions or commands

oai_token-act-pair · gpt-3.5-turbo

symbols or punctuation marks that convey emotional or expressive nuances

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 unveiled

-0.69

velop

-0.63

 dispatched

-0.61

eport

-0.55

senal

-0.55

iled

-0.55

 neighbourhood

-0.54

 Released

-0.54

 flagged

-0.54

 besieged

-0.54

POSITIVE LOGITS

whatever

1.20

yeah

1.18

unless

1.17

anything

1.15

etc

1.15

laughs

1.12

 blah

1.10

 haha

1.04

but

1.03

etc

1.03

Activations Density 0.484%

phrases related to instructions or commands

symbols or punctuation marks that convey emotional or expressive nuances

No Comments

No Known Activations