INDEX

Explanations

names and terms related to individuals

oai_token-act-pair · gpt-3.5-turbo

variations of the name "Harris" and related terms

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

 contrace

-0.77

İĭ

-0.73

©¶æ

-0.72

ortium

-0.70

 subp

-0.67

orius

-0.66

mosp

-0.65

 coord

-0.62

rafted

-0.60

mber

-0.59

POSITIVE LOGITS

SHA

0.80

aughs

0.76

raid

0.73

schild

0.70

ette

0.70

inger

0.64

zee

0.64

ios

0.63

ROR

0.62

inton

0.62

Activations Density 0.140%

names and terms related to individuals

variations of the name "Harris" and related terms

No Comments

No Known Activations