INDEX

Explanations

terms related to building and constructing systems or frameworks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

nie

-0.16

itar

-0.15

ureau

-0.15

å½

-0.15

åĩĿ

-0.15

igator

-0.14

agg

-0.14

entric

-0.14

 ÏĥÎºÎ¿

-0.14

zzo

-0.14

POSITIVE LOGITS

oller

0.17

@qq

0.16

ODULE

0.14

abase

0.14

Ã¡Äį

0.14

atat

0.13

eydi

0.13

eldo

0.13

uzzi

0.13

erville

0.13

Activations Density 0.092%

terms related to building and constructing systems or frameworks

No Comments

No Known Activations