INDEX

Explanations

references to technology and specific products or companies related to technology

oai_token-act-pair · gpt-3.5-turbo

references to specific nuclear reactor codes and related terminology

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

ovych

-0.77

idays

-0.73

schild

-0.62

angible

-0.60

 tremend

-0.58

gat

-0.58

inen

-0.57

 sounding

-0.57

 spinning

-0.55

uku

-0.54

POSITIVE LOGITS

ulhu

1.09

arette

0.79

pillar

0.77

ioxide

0.75

ãĤ¼ãĤ¦ãĤ¹

0.71

emporary

0.70

Psy

0.69

 Dating

0.67

berus

0.67

cled

0.67

Activations Density 0.174%

references to technology and specific products or companies related to technology

references to specific nuclear reactor codes and related terminology

No Comments

No Known Activations