INDEX

Explanations

references to events, programs, or organized activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

 Humph

-0.15

_LINEAR

-0.15

 Rosenstein

-0.14

ideon

-0.14

ABCDEFGHIJKLMNOP

-0.14

 Huff

-0.14

lda

-0.13

styleType

-0.13

ALSE

-0.13

oda

-0.13

POSITIVE LOGITS

quia

0.17

ibase

0.15

adow

0.15

jax

0.15

.scalablytyped

0.15

ees

0.15

ÐµÐ±

0.14

dato

0.14

mand

0.13

mmo

0.13

Activations Density 0.178%

references to events, programs, or organized activities

No Comments

No Known Activations