INDEX

Explanations

phrases expressing uncertainty or hypothetical situations

oai_token-act-pair · gpt-3.5-turbo

phrases that express uncertainty or hypothetical situations

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

 nomine

-0.75

Sys

-0.68

 Manufacturer

-0.66

nec

-0.63

ãĥ³ãĤ¸

-0.63

WB

-0.61

 seldom

-0.61

----------------------------------------------------------------

-0.61

phy

-0.60

ÃįÃį

-0.60

POSITIVE LOGITS

 doubt

0.79

 doubts

0.77

 develops

0.75

 else

0.75

 bothers

0.69

 doubted

0.67

 happens

0.67

 wonders

0.65

 someday

0.65

 ever

0.65

Activations Density 0.053%

phrases expressing uncertainty or hypothetical situations

phrases that express uncertainty or hypothetical situations

No Comments

No Known Activations