INDEX

Explanations

certain key identifiers and metadata associated with articles or reports

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

uada

-0.16

agen

-0.15

-0.14

 Rough

-0.14

vig

-0.14

 Cove

-0.14

å°¼äºļ

-0.14

Ø¶Ø§ÙĨ

-0.14

igar

-0.14

ecz

-0.14

POSITIVE LOGITS

ouro

0.16

variants

0.15

oice

0.15

addock

0.15

.Inner

0.15

Äģn

0.15

uter

0.14

 graduation

0.14

Responder

0.14

har

0.14

Activations Density 0.018%

certain key identifiers and metadata associated with articles or reports

No Comments

No Known Activations