INDEX

Explanations

mentions of government positions and departments

oai_token-act-pair · gpt-3.5-turbo

references to government officials or positions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Coverage

-1.15

 Balls

-1.12

 Piece

-1.08

 Scores

-1.08

 Areas

-1.08

 Claim

-1.07

 Parties

-1.07

 Pressure

-1.05

 Amount

-1.04

 Fans

-1.03

POSITIVE LOGITS

 senate

1.10

 commons

1.05

 sciences

1.05

 treasurer

0.96

 chancellor

0.95

 agriculture

0.94

 queen

0.93

 corps

0.91

 arts

0.91

 americ

0.90

Activations Density 0.514%

mentions of government positions and departments

references to government officials or positions

No Comments

No Known Activations