INDEX

Explanations

terms related to advancement or improvement 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 @ 3-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L3R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.3.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

ethyst

-0.15

abei

-0.14

reon

-0.14

odka

-0.14

bre

-0.14

 unset

-0.13

offee

-0.13

sein

-0.13

 cocci

-0.13

apia

-0.13

POSITIVE LOGITS

bilt

0.18

ve

0.17

ions

0.16

ional

0.15

ion

0.15

filt

0.15

æľ

0.15

FromBody

0.15

ãĤ¦

0.14

rar

0.14

Activations Density 0.015%

terms related to advancement or improvement in various contexts

No Comments

No Known Activations