INDEX

Explanations

proper nouns related to people's names or titles

oai_token-act-pair · gpt-3.5-turbo

occurrences of the substring "oc" in words

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

Zed

-0.68

DERR

-0.68

CPC

-0.63

growth

-0.62

myra

-0.61

vertisement

-0.61

 visitation

-0.59

âĶĢâĶĢ

-0.59

nings

-0.58

DIT

-0.56

POSITIVE LOGITS

hester

0.86

hent

0.82

hett

0.77

hen

0.72

atoon

0.70

omo

0.69

umbers

0.69

eny

0.67

henko

0.66

ap

0.66

Activations Density 0.117%

proper nouns related to people's names or titles

occurrences of the substring "oc" in words

No Comments

No Known Activations