INDEX

Explanations

mentions related to governmental actions or structures

oai_token-act-pair · gpt-3.5-turbo

mentions of government or state entities and their actions

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

poons

-0.93

aukee

-0.80

poon

-0.79

veyard

-0.77

ayson

-0.77

byss

-0.73

CAST

-0.73

odi

-0.73

 APPLIC

-0.72

regor

-0.72

POSITIVE LOGITS

rooms

0.99

lessness

0.91

less

0.85

hood

0.83

room

0.81

 apparatus

0.80

wide

0.77

 monopoly

0.76

 legislatures

0.75

 relations

0.75

Activations Density 0.054%

mentions related to governmental actions or structures

mentions of government or state entities and their actions

No Comments

No Known Activations