INDEX

Explanations

adjectives describing quality and value

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

ÑģÑĤÑĢÐ¸

-0.15

íĥĪ

-0.15

adh

-0.14

/Dk

-0.14

olum

-0.14

 èĩ³

-0.14

naments

-0.13

IRC

-0.13

stial

-0.13

ippy

-0.13

POSITIVE LOGITS

-looking

0.20

 stuff

0.16

argar

0.15

.getOwnProperty

0.14

oldt

0.14

 things

0.14

aha

0.13

Æ°

0.13

Activations Density 0.032%

adjectives describing quality and value

No Comments

No Known Activations