INDEX

Explanations

expressions indicating user familiarity and adaptation to new experiences or systems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

 Rubin

-0.16

memberOf

-0.16

asha

-0.15

ê³¡

-0.14

olec

-0.14

ules

-0.14

bor

-0.14

elper

-0.14

uye

-0.13

ylon

-0.13

POSITIVE LOGITS

 used

0.85

 Used

0.79

Used

0.75

used

0.74

 USED

0.70

-used

0.63

_used

0.61

.used

0.60

USED

0.60

 accustomed

0.57

Activations Density 0.184%

expressions indicating user familiarity and adaptation to new experiences or systems

No Comments

No Known Activations