INDEX

Explanations

names related to the character "Max" or phrases with the word "axe"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the substring "ax" within words

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

perature

-0.67

ishable

-0.63

ja

-0.62

isSpecialOrderable

-0.60

 Bethesda

-0.60

 Shades

-0.60

âĺħâĺħ

-0.59

ufact

-0.59

ETHOD

-0.58

FTWARE

-0.58

POSITIVE LOGITS

xon

1.25

xus

1.17

es

1.00

seed

0.95

xes

0.92

iang

0.91

endale

0.90

illary

0.88

ercise

0.87

posure

0.86

Activations Density 0.020%

names related to the character "Max" or phrases with the word "axe"

occurrences of the substring "ax" within words

No Comments

No Known Activations