INDEX

Explanations

the word "quite" 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 @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

asca

-0.16

izar

-0.16

ì°¬

-0.15

UEL

-0.15

Ø§Ø¸

-0.15

ãĥ³ãĥĨ

-0.14

 Bien

-0.14

istema

-0.14

 scatter

-0.14

uel

-0.14

POSITIVE LOGITS

Ĥæķ°

0.17

/stdc

0.16

 INTERRUPTION

0.15

omap

0.14

ernity

0.14

çĻ

0.14

peq

0.14

.dtd

0.13

skin

0.13

asions

0.13

Activations Density 0.015%

the word "quite" in various contexts

No Comments

No Known Activations