INDEX

Explanations

words related to new information or research findings

oai_token-act-pair · gpt-3.5-turbo

phrases that reference new reports or studies

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

respective

-0.77

urations

-0.77

 preceded

-0.74

 secondly

-0.70

 reverted

-0.69

Reloaded

-0.69

cius

-0.68

ecause

-0.68

ochond

-0.66

Ò

-0.66

POSITIVE LOGITS

 report

1.07

 study

1.06

study

1.00

 infographic

1.00

 documentary

0.97

 analysis

0.95

 poll

0.95

 survey

0.95

 biography

0.91

bie

0.91

Activations Density 0.079%

words related to new information or research findings

phrases that reference new reports or studies

No Comments

No Known Activations