INDEX

Explanations

references to intellectual or artistic output

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

roman

-0.16

ÑĭÑĪ

-0.15

uned

-0.14

bra

-0.14

FETCH

-0.14

æľĭ

-0.14

alars

-0.14

edException

-0.14

-desc

-0.14

 doomed

-0.14

POSITIVE LOGITS

ear

0.16

swer

0.15

onet

0.15

JsonValue

0.15

tow

0.14

ouser

0.14

/***************************************************************************↵

0.14

Ä±lÄ±ÅŁ

0.14

ãĥ«ãĥĪ

0.14

 MatTable

0.13

Activations Density 0.385%

references to intellectual or artistic output

No Comments

No Known Activations