INDEX

Explanations

the word "can" in various contexts indicating ability or options

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

airo

-0.18

stvo

-0.15

ignon

-0.15

aurus

-0.14

etto

-0.14

 itself

-0.14

ÙĨØ¯ÙĬ

-0.14

irror

-0.14

ngo

-0.13

undry

-0.13

POSITIVE LOGITS

 freely

0.18

0.16

’t

0.15

elas

0.15

0.14

acre

0.14

cheng

0.14

LEE

0.14

Activations Density 0.077%

the word "can" in various contexts indicating ability or options

No Comments

No Known Activations