INDEX

Explanations

occurrences of the abbreviation "sp" or related terms about special features or specifications 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

antry

-0.18

ãĥ¼ãĤ

-0.15

emple

-0.15

bsolute

-0.15

ictory

-0.14

uct

-0.14

vous

-0.14

.AddParameter

-0.14

 exponent

-0.14

isper

-0.14

POSITIVE LOGITS

jang

0.15

ERO

0.14

 trop

0.14

roz

0.14

cum

0.14

argo

0.14

.setAction

0.14

 Bruno

0.14

fa

0.13

Activations Density 0.024%

occurrences of the abbreviation "sp" or related terms about special features or specifications in various contexts

No Comments

No Known Activations