INDEX

Explanations

the word "prone" and related concepts

oai_token-act-pair · gpt-3.5-turbo

phrases indicating susceptibility or likelihood of occurrence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

tein

-0.88

flies

-0.75

 CLSID

-0.67

ARE

-0.66

=-=-=-=-

-0.65

GROUP

-0.65

 VIDEOS

-0.64

mson

-0.64

NZ

-0.62

zos

-0.62

POSITIVE LOGITS

theless

0.94

entimes

0.94

 prone

0.81

eus

0.79

 confir

0.78

ritis

0.77

itability

0.73

untarily

0.71

 embr

0.71

icter

0.70

Activations Density 0.032%

the word "prone" and related concepts

phrases indicating susceptibility or likelihood of occurrence

No Comments

No Known Activations