INDEX

Explanations

terms related to software versioning and compatibility

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

ardo

-0.15

 ÙħÙĤØ¯

-0.15

 Moody

-0.15

 Rubin

-0.15

pter

-0.14

oky

-0.14

 kino

-0.14

ADS

-0.14

 Levin

-0.13

ekim

-0.13

POSITIVE LOGITS

igaret

0.15

inson

0.14

lit

0.14

nod

0.13

orph

0.13

igned

0.13

.openg

0.13

iscard

0.13

ÙĪÛĮØ¯

0.13

nar

0.13

Activations Density 0.105%

terms related to software versioning and compatibility

No Comments

No Known Activations