INDEX

Explanations

proper names, specifically the name "Hans."

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Hans" and related characters

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

aneously

-0.92

 Scotia

-0.87

 IMAGES

-0.80

otions

-0.72

cial

-0.71

ista

-0.69

bably

-0.67

aneous

-0.67

zx

-0.67

ersion

-0.66

POSITIVE LOGITS

hin

1.10

sein

1.05

iewicz

0.86

-+

0.85

 Andersen

0.82

hu

0.81

hus

0.81

 Hans

0.81

 Zimmer

0.77

von

0.77

Activations Density 0.015%

proper names, specifically the name "Hans."

mentions of the name "Hans" and related characters

No Comments

No Known Activations