INDEX

Explanations

instances of the letter 'l' 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 @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

appa

-0.17

ake

-0.17

Lux

-0.16

idis

-0.14

indre

-0.14

igner

-0.14

uco

-0.14

ama

-0.14

ite

-0.14

Lux

-0.13

POSITIVE LOGITS

lew

0.19

aws

0.17

ieber

0.17

uria

0.16

obb

0.16

Aw

0.16

rael

0.16

est

0.16

iable

0.16

ured

0.16

Activations Density 0.032%

instances of the letter 'l' in various contexts

No Comments

No Known Activations