INDEX

Explanations

sentences related to ideology and beliefs

oai_token-act-pair · gpt-3.5-turbo

political language surrounding issues of manipulation and power dynamics

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

DragonMagazine

-0.69

iple

-0.65

 Originally

-0.61

 Deadline

-0.60

raft

-0.59

©¶æ¥µ

-0.58

 Warehouse

-0.57

ortium

-0.56

availability

-0.56

 Medline

-0.55

POSITIVE LOGITS

 themselves

0.80

 subord

0.77

 ignor

0.77

 immoral

0.73

 unworthy

0.70

 inconvenient

0.70

 legitim

0.70

 bigotry

0.70

 ignorant

0.69

 undue

0.69

Activations Density 1.119%

sentences related to ideology and beliefs

political language surrounding issues of manipulation and power dynamics

No Comments

No Known Activations