INDEX

Explanations

conditional phrases indicating uncertainty or possibility

oai_token-act-pair · gpt-3.5-turbo

conditional phrases that express uncertainty or possibilities

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 tremend

-0.67

azon

-0.60

ulic

-0.56

ãĤ³

-0.54

ullivan

-0.53

anto

-0.53

ouver

-0.53

ino

-0.52

OX

-0.52

imo

-0.51

POSITIVE LOGITS

iability

0.61

 anything

0.61

stice

0.60

ansk

0.59

any

0.59

alties

0.59

leans

0.58

shed

0.54

ĨĴ

0.54

ists

0.54

Activations Density 0.142%

conditional phrases indicating uncertainty or possibility

conditional phrases that express uncertainty or possibilities

No Comments

No Known Activations