INDEX

Explanations

terms related to health metrics and comparisons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

edException

-0.17

ettle

-0.16

ocab

-0.15

ombok

-0.15

.radio

-0.15

 Feinstein

-0.14

dio

-0.14

å°¼äºļ

-0.14

Ð°Ð½Ð¸ÑĨ

-0.14

Subsystem

-0.14

POSITIVE LOGITS

odesk

0.16

claimer

0.15

ainter

0.15

inÃ¡

0.15

 Crystal

0.15

enco

0.14

ALA

0.14

ernaut

0.14

Activations Density 0.007%

terms related to health metrics and comparisons

No Comments

No Known Activations