INDEX

Explanations

phrases that denote roles or functions in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

èĪª

-0.16

asurer

-0.16

elter

-0.15

alez

-0.15

ÙĪØªØ±

-0.15

åŁºåľ°

-0.15

ignum

-0.15

ogra

-0.14

strt

-0.14

ilig

-0.14

POSITIVE LOGITS

 Copy

0.15

 copy

0.15

 Gall

0.14

Er

0.14

fm

0.14

astos

0.14

 doÄŁ

0.14

ÙĬØ§Øª

0.14

asto

0.14

onom

0.14

Activations Density 0.014%

phrases that denote roles or functions in various contexts

No Comments

No Known Activations