INDEX

Explanations

political entities, individuals, or entities relevant in various fields or contexts

oai_token-act-pair · gpt-3.5-turbo

references to specific groups and their roles in various contexts

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

ategory

-0.86

EStreamFrame

-0.82

arnaev

-0.79

srf

-0.77

eanor

-0.72

entanyl

-0.69

rontal

-0.69

raints

-0.68

aughtered

-0.68

 carbohyd

-0.67

POSITIVE LOGITS

 deems

0.97

 lacked

0.78

 sorely

0.77

 lacks

0.76

 might

0.75

 deem

0.72

 possesses

0.67

 excel

0.67

 wished

0.67

 ought

0.66

Activations Density 0.449%

political entities, individuals, or entities relevant in various fields or contexts

references to specific groups and their roles in various contexts

No Comments

No Known Activations