INDEX

Explanations

the name "Malzahn" or similar variations

oai_token-act-pair · gpt-3.5-turbo

mentions of specific names or terms related to "Malzahn."

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

:{

-0.81

Reviewer

-0.74

 horm

-0.70

 Bethesda

-0.69

kefeller

-0.66

 explan

-0.66

bley

-0.66

 magnification

-0.62

Fine

-0.62

glers

-0.60

POSITIVE LOGITS

zan

0.87

amia

0.80

ady

0.73

ubi

0.72

adin

0.72

ieri

0.72

aughs

0.72

opol

0.71

acus

0.70

ãĤ¶

0.70

Activations Density 0.099%

the name "Malzahn" or similar variations

mentions of specific names or terms related to "Malzahn."

No Comments

No Known Activations