INDEX

Explanations

terms related to scale and size in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

lice

-0.16

unt

-0.15

 latter

-0.15

unch

-0.15

elay

-0.14

lok

-0.14

.ptr

-0.14

odel

-0.14

iaz

-0.13

ew

-0.13

POSITIVE LOGITS

/high

0.17

edly

0.16

iros

0.15

urance

0.14

/Branch

0.14

ÃľM

0.14

NibName

0.14

LD

0.14

feb

0.14

icism

0.14

Activations Density 0.159%

terms related to scale and size in various contexts

No Comments

No Known Activations

terms related to scale and size in various contexts

No Comments

No Known Activations