INDEX

Explanations

phrases related to government directives and policies

oai_token-act-pair · gpt-3.5-turbo

references to government directives and agency actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

jew

-0.89

 Whats

-0.80

 organisers

-0.78

 favourite

-0.78

 neighbour

-0.77

 colourful

-0.77

 paradise

-0.76

 Laugh

-0.76

 neighbours

-0.75

hur

-0.75

POSITIVE LOGITS

HHS

1.18

Recommend

1.14

 adequ

1.12

 Recommend

1.11

 guidance

1.10

 recommendations

1.10

 Implementation

1.09

 promul

1.08

 promulg

1.07

 determin

1.06

Activations Density 0.660%

phrases related to government directives and policies

references to government directives and agency actions

No Comments

No Known Activations