INDEX

Explanations

references to conditions and criteria that vary based on specific situations or factors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

imens

-0.19

istr

-0.18

ifen

-0.16

ware

-0.15

inya

-0.15

tones

-0.15

imin

-0.14

chen

-0.14

ifu

-0.14

iba

-0.14

POSITIVE LOGITS

à¥Įà¤²

0.15

heim

0.15

 whether

0.14

enth

0.14

age

0.14

ÐºÑĥÐ»

0.13

 type

0.13

 degree

0.13

cul

0.13

fen

0.13

Activations Density 0.030%

references to conditions and criteria that vary based on specific situations or factors

No Comments

No Known Activations