INDEX

Explanations

technical specifications and programming context

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

hir

-0.16

ãģĬ

-0.16

dit

-0.15

angu

-0.14

ovo

-0.14

ÑĩÑĥ

-0.14

.travel

-0.14

ì²ł

-0.14

 cocci

-0.14

ILT

-0.14

POSITIVE LOGITS

ackbar

0.17

SCAN

0.16

MK

0.14

iaux

0.14

ration

0.14

ibern

0.14

utton

0.14

inar

0.14

tout

0.14

 cast

0.14

Activations Density 0.387%

technical specifications and programming context

No Comments

No Known Activations