INDEX

Explanations

the word "All" in various contexts and forms

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

icons

-0.17

icon

-0.16

empt

-0.16

icky

-0.15

ilion

-0.15

.club

-0.15

osa

-0.14

illions

-0.14

Two

-0.14

icc

-0.14

POSITIVE LOGITS

ergic

0.19

iances

0.19

 Souls

0.19

erdings

0.19

iance

0.18

 Terrain

0.18

URED

0.18

ignment

0.17

geme

0.17

erg

0.17

Activations Density 0.034%

the word "All" in various contexts and forms

No Comments

No Known Activations