INDEX

Explanations

references to data and statistical metrics in research contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

ÏģÏħ

-0.16

lech

-0.16

(--

-0.15

Interstitial

-0.14

[++

-0.14

ansi

-0.14

 ì¡

-0.14

_MULT

-0.14

 tolua

-0.14

incy

-0.14

POSITIVE LOGITS

urm

0.16

 balance

0.15

 clock

0.14

sw

0.14

↵

0.14

rem

0.14

omen

0.14

 Gott

0.13

UREMENT

0.13

AP

0.13

Activations Density 0.244%

references to data and statistical metrics in research contexts

No Comments

No Known Activations