INDEX

Explanations

complex phrases and structures involving nouns and descriptors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 Recognition

-0.15

hangi

-0.15

aston

-0.15

VertexAttrib

-0.14

Recognition

-0.14

ongo

-0.13

ais

-0.13

ERRU

-0.13

loom

-0.13

alg

-0.13

POSITIVE LOGITS

ipple

0.18

odal

0.14

ç¬¦

0.14

anking

0.14

ëĥ¥

0.14

Ð½ÐµÐ´

0.14

_SUITE

0.13

omed

0.13

Activations Density 0.024%

complex phrases and structures involving nouns and descriptors

No Comments

No Known Activations