INDEX

Explanations

characters indicating strong emotional responses or reactions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

ÃŃnh

-0.17

.INSTANCE

-0.15

rello

-0.15

å¥¹ä»¬

-0.14

 Morr

-0.14

$MESS

-0.14

rescia

-0.14

icari

-0.14

\Validation

-0.14

quee

-0.14

POSITIVE LOGITS

 tome

0.16

 interrupt

0.15

 paramMap

0.14

 schle

0.14

 seating

0.14

 bustling

0.14

zas

0.14

 dilig

0.14

interrupt

0.14

 peek

0.13

Activations Density 0.001%

characters indicating strong emotional responses or reactions

No Comments

No Known Activations