INDEX

Explanations

statements with uncertainty or speculation

oai_token-act-pair · gpt-3.5-turbo

phrases related to evidence and claims in discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

inav

-0.65

 pione

-0.64

ãĥĩãĤ£

-0.62

arse

-0.62

ãĤ¨ãĥ«

-0.62

 subur

-0.61

ãĤ´ãĥ³

-0.60

ãĤ¼ãĤ¦ãĤ¹

-0.57

ãĥĵ

-0.57

è»

-0.55

POSITIVE LOGITS

it

0.57

 Americans

0.56

 anyone

0.54

 individuals

0.54

 Britons

0.54

 tensions

0.53

 improper

0.53

 landlords

0.52

 people

0.51

 certain

0.51

Activations Density 0.912%

statements with uncertainty or speculation

phrases related to evidence and claims in discussions

No Comments

No Known Activations