INDEX

Explanations

occurrences of the character 'i' 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

ulner

-0.15

Pry

-0.15

otti

-0.15

vise

-0.15

 Permanent

-0.15

aldi

-0.15

´Ī

-0.15

act

-0.15

POSITIVE LOGITS

xfff

0.15

StateToProps

0.15

oÄŁ

0.14

-mf

0.14

Ã©ru

0.14

_pb

0.14

icus

0.14

_NS

0.13

zed

0.13

ãģ£ãģ

0.13

Activations Density 0.036%

occurrences of the character 'i' in various contexts

No Comments

No Known Activations