INDEX

Explanations

phrases related to news articles or reports, specifically those with dates and sources

oai_token-act-pair · gpt-3.5-turbo

occurrences of date and location references

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 prol

-0.61

 next

-0.60

worms

-0.60

 fame

-0.60

 attent

-0.58

 conversion

-0.58

onics

-0.58

 Spartans

-0.57

 Sixth

-0.57

 brakes

-0.57

POSITIVE LOGITS

Jonathan

1.05

String

1.03

Jason

0.99

David

0.99

Mike

0.98

Carl

0.97

Joshua

0.95

Jim

0.95

Luc

0.92

Kevin

0.92

Activations Density 0.021%

phrases related to news articles or reports, specifically those with dates and sources

occurrences of date and location references

No Comments

No Known Activations