INDEX

Explanations

occurrences of punctuation and card-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

zie

-0.16

éĸĢ

-0.16

iou

-0.16

Paz

-0.15

enc

-0.15

 Schwartz

-0.14

ople

-0.14

ension

-0.14

 Commonwealth

-0.14

Las

-0.14

POSITIVE LOGITS

ROY

0.17

pez

0.15

loys

0.14

Ð¸Ð±

0.14

baby

0.14

arrant

0.14

aucoup

0.14

Ð¸Ð½Ð³

0.14

ETO

0.14

_exempt

0.14

Activations Density 0.006%

occurrences of punctuation and card-related terms

No Comments

No Known Activations