INDEX

Explanations

references to programs, approvals, and debates

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

Wel

-0.16

kos

-0.15

çħ

-0.15

Ä¾

-0.15

 Corinth

-0.15

inki

-0.14

ÑıÑĤÑĮ

-0.14

Pis

-0.13

sil

-0.13

etable

-0.13

POSITIVE LOGITS

compound

0.16

ÃĹ↵↵

0.15

,eg

0.15

ibold

0.15

aepernick

0.15

assa

0.15

RequiredMixin

0.15

erdem

0.15

unami

0.15

Activations Density 0.034%

references to programs, approvals, and debates

No Comments

No Known Activations