INDEX

Explanations

phrases indicating the beginning of new thoughts or sections in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

ripper

-0.15

Haw

-0.15

aria

-0.14

rÃ³

-0.14

quete

-0.14

ngine

-0.14

ayo

-0.14

arial

-0.14

ants

-0.14

ched

-0.14

POSITIVE LOGITS

/Form

0.15

ãģªãĤĭ

0.14

å¥ĩ

0.14

onn

0.14

iage

0.14

sted

0.14

atar

0.14

ovice

0.14

etimes

0.14

.ManyToMany

0.14

Activations Density 0.032%

phrases indicating the beginning of new thoughts or sections in the text

No Comments

No Known Activations