INDEX

Explanations

references to the term "Haus" with varying levels of activation

oai_token-act-pair · gpt-3.5-turbo

mentions of a specific entity or brand with the term "Haus."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

âĶĢâĶĢ

-0.71

td

-0.70

OUT

-0.63

 cane

-0.62

 short

-0.62

 differentiated

-0.59

 pleasure

-0.58

cue

-0.58

Assembly

-0.57

 flashes

-0.57

POSITIVE LOGITS

chwitz

1.16

ername

1.03

kas

1.02

uin

0.95

aus

0.94

aurus

0.91

ylum

0.91

atell

0.88

adian

0.86

perm

0.83

Activations Density 0.018%

references to the term "Haus" with varying levels of activation

mentions of a specific entity or brand with the term "Haus."

No Comments

No Known Activations