INDEX

Explanations

medical terms related to procedures or conditions

oai_token-act-pair · gpt-3.5-turbo

abbreviations and truncated words

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ģĸ

-0.67

hyde

-0.66

HAEL

-0.66

 LORD

-0.63

 Darling

-0.63

å§«

-0.62

 Duchess

-0.62

 Solution

-0.62

NAD

-0.61

 Sorceress

-0.60

POSITIVE LOGITS

ag

0.91

umb

0.90

ashing

0.89

aring

0.89

ogging

0.88

icy

0.87

acet

0.86

anging

0.86

idd

0.86

ars

0.86

Activations Density 0.449%

medical terms related to procedures or conditions

abbreviations and truncated words

No Comments

No Known Activations