INDEX

Explanations

uncertainty or lack of knowledge, especially in the form of questions

oai_token-act-pair · gpt-3.5-turbo

expressions of uncertainty or lack of knowledge

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ueller

-0.71

 exting

-0.71

oided

-0.66

ammers

-0.65

wagen

-0.64

stro

-0.63

ccording

-0.62

aez

-0.62

onding

-0.61

tein

-0.61

POSITIVE LOGITS

why

1.34

how

1.32

 whether

1.27

 what

1.19

if

1.16

 anything

1.06

 anymore

1.04

WHY

1.04

why

1.01

 exactly

1.01

Activations Density 0.041%

uncertainty or lack of knowledge, especially in the form of questions

expressions of uncertainty or lack of knowledge

No Comments

No Known Activations