INDEX

Explanations

structured data formats, particularly JSON objects and elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

tro

-0.17

atra

-0.16

tro

-0.15

ngle

-0.14

tape

-0.14

ens

-0.14

Ø§Ø³Ø±

-0.14

eur

-0.14

akh

-0.14

uria

-0.14

POSITIVE LOGITS

Ja

0.16

itzer

0.15

ëį°ìĿ´íĬ¸

0.14

à¸¡à¸Ń

0.14

HITE

0.14

sep

0.13

swire

0.13

ungi

0.13

clidean

0.13

Veg

0.13

Activations Density 0.249%

structured data formats, particularly JSON objects and elements

No Comments

No Known Activations