INDEX

Explanations

references to controller elements in code snippets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

teenth

-0.18

ening

-0.17

istr

-0.16

nes

-0.15

dal

-0.15

ings

-0.15

æļ®

-0.15

oble

-0.15

rlen

-0.15

Äįas

-0.15

POSITIVE LOGITS

ial

0.16

/action

0.16

zeit

0.14

 positively

0.14

uang

0.14

*pow

0.14

esp

0.14

ateral

0.14

-Level

0.14

ural

0.14

Activations Density 0.012%

references to controller elements in code snippets

No Comments

No Known Activations