INDEX

Explanations

words related to city governance and related activities such as approval, voting, and meetings

oai_token-act-pair · gpt-3.5-turbo

references to city council activities and decisions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

sis

-0.75

stream

-0.72

eworld

-0.70

ihad

-0.70

flight

-0.69

ihadi

-0.68

ylan

-0.68

uddin

-0.66

nikov

-0.66

ython

-0.65

POSITIVE LOGITS

 district

0.96

 commissioners

0.93

 districts

0.89

 zoning

0.89

Council

0.86

 commissioner

0.82

 plaza

0.80

 council

0.79

 runoff

0.78

 ordinances

0.77

Activations Density 0.267%

words related to city governance and related activities such as approval, voting, and meetings

references to city council activities and decisions

No Comments

No Known Activations