INDEX

Explanations

words related to medical conditions or terms related to the medical field

oai_token-act-pair · gpt-3.5-turbo

occurrences of the substring "um" regardless of context

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

¬¼

-0.65

externalActionCode

-0.65

 strawberries

-0.64

IGHTS

-0.63

 resemb

-0.62

 DARK

-0.61

 unaff

-0.60

RAW

-0.59

 Marketable

-0.59

 impunity

-0.59

POSITIVE LOGITS

mers

1.20

ming

1.15

pty

1.14

etric

0.97

cised

0.97

ptive

0.96

atism

0.94

essage

0.93

ueller

0.89

ulative

0.89

Activations Density 0.049%

words related to medical conditions or terms related to the medical field

occurrences of the substring "um" regardless of context

No Comments

No Known Activations