INDEX

Explanations

the letter 'd' 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 @ 1-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

iversite

-0.16

rÃ¡

-0.15

oles

-0.15

semble

-0.15

ÙĪÙĦÙĬ

-0.15

dden

-0.14

yg

-0.14

.opend

-0.14

á»Ļn

-0.14

ogs

-0.14

POSITIVE LOGITS

fen

0.17

tax

0.14

Aqu

0.14

 rank

0.14

 earlier

0.14

atori

0.14

harma

0.14

int

0.14

 spell

0.14

ation

0.13

Activations Density 0.025%

the letter 'd' in various contexts

No Comments

No Known Activations