INDEX

Explanations

mentions of geopolitical locations or political entities

oai_token-act-pair · gpt-3.5-turbo

terms and phrases related to public issues and discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

osate

-0.73

ngth

-0.67

oÄŁ

-0.62

ãĥ´

-0.62

bourg

-0.61

ummer

-0.59

uesday

-0.58

inen

-0.58

ensen

-0.58

oulos

-0.57

POSITIVE LOGITS

Appearances

0.75

ience

0.70

ī

0.70

º

0.68

ions

0.68

IENT

0.67

entials

0.66

isphere

0.65

itus

0.64

ional

0.64

Activations Density 0.132%

mentions of geopolitical locations or political entities

terms and phrases related to public issues and discussions

No Comments

No Known Activations