INDEX

Explanations

affirmations and emphatic expressions within conversational or narrative contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ses

-0.19

ctors

-0.15

efeller

-0.15

ssp

-0.15

ipi

-0.14

èĪ¬

-0.14

pires

-0.14

pesan

-0.14

avings

-0.14

ese

-0.14

POSITIVE LOGITS

oret

0.30

orem

0.24

oretical

0.23

ories

0.21

give

0.18

semble

0.18

0.17

suite

0.16

ronic

0.16

maal

0.14

Activations Density 2.224%

affirmations and emphatic expressions within conversational or narrative contexts

No Comments

No Known Activations