INDEX

Explanations

references to various museums and states

oai_token-act-pair · gpt-3.5-turbo

references to state government entities and museums

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

 Characters

-1.09

 Spread

-1.08

 Parties

-1.08

 Treatment

-1.07

 Expression

-1.06

 Habit

-1.06

 Crew

-1.04

 Balls

-1.04

 Background

-1.04

 Opposition

-1.04

POSITIVE LOGITS

 americ

1.24

 english

1.05

 aust

1.05

 senate

1.05

 commons

1.03

 chancellor

1.01

 queen

1.00

 king

0.98

 knight

0.95

 princ

0.95

Activations Density 0.570%

references to various museums and states

references to state government entities and museums

No Comments

No Known Activations