INDEX

Explanations

titles of television shows and films

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

¾

-0.18

ÑĥÑĢÐ½

-0.16

ãĥ³ãĥĶ

-0.16

iagnostics

-0.15

kova

-0.14

odont

-0.14

issan

-0.14

iras

-0.14

Ø²Ø§Ùħ

-0.14

oved

-0.14

POSITIVE LOGITS

observeOn

0.16

ä»ģ

0.15

ome

0.14

oll

0.14

rex

0.14

ardon

0.14

 classical

0.13

Nes

0.13

 Kemp

0.13

Activations Density 0.054%

titles of television shows and films

No Comments

No Known Activations