INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

terms related to frustration or dissatisfaction

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ryu

-0.80

una

-0.69

cott

-0.69

nz

-0.69

*)

-0.69

*)

-0.68

hire

-0.68

kell

-0.68

-0.67

nm

-0.65

POSITIVE LOGITS

onomy

0.69

Pwr

0.68

itol

0.65

 Shed

0.65

LAW

0.63

icultural

0.62

Hat

0.61

assis

0.58

 suscept

0.58

ogeneity

0.57

Activations Density 0.000%

proper nouns

terms related to frustration or dissatisfaction

No Comments

No Known Activations