INDEX

Explanations

the word "any" followed by numerical values or specific actions

oai_token-act-pair · gpt-3.5-turbo

the word "any" in various contexts

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

rex

-0.81

expensive

-0.78

rez

-0.73

seless

-0.69

bows

-0.65

comes

-0.65

rea

-0.63

rend

-0.63

efficient

-0.63

yp

-0.63

POSITIVE LOGITS

THING

1.31

WHERE

0.97

 conceivable

0.94

body

0.87

 kind

0.84

ONE

0.82

thin

0.82

 sort

0.79

where

0.79

 semblance

0.79

Activations Density 0.074%

the word "any" followed by numerical values or specific actions

the word "any" in various contexts

No Comments

No Known Activations