INDEX

Explanations

numeric or statistical information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

Ã³j

-0.18

alin

-0.16

alic

-0.15

 Ð¿ÑĢÐµÐ´Ð²Ð°ÑĢ

-0.15

andi

-0.15

lez

-0.15

Ä±lÄ±Ã§

-0.14

 McCartney

-0.14

 hann

-0.14

çĸ²

-0.14

POSITIVE LOGITS

erk

0.16

oud

0.15

uest

0.14

antha

0.14

.strict

0.14

essel

0.14

 Mans

0.14

ReturnValue

0.13

esome

0.13

Activations Density 0.037%

numeric or statistical information

No Comments

No Known Activations