INDEX

Explanations

instances of the word "Wel" and its variations, highlighting a focus on specific names or labels

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

 Rich

-0.15

 Chief

-0.14

 toes

-0.14

ruits

-0.14

arda

-0.14

aped

-0.14

hood

-0.14

ml

-0.14

eo

-0.14

tom

-0.14

POSITIVE LOGITS

ROUGH

0.18

0.16

kyt

0.16

ienne

0.16

anglicky

0.15

ÃŃÅ¡

0.15

sez

0.15

bedo

0.15

Ð½ÐµÑĢÐ³

0.14

Crud

0.14

Activations Density 0.011%

instances of the word "Wel" and its variations, highlighting a focus on specific names or labels

No Comments

No Known Activations