INDEX

Explanations

words related to "mar" and "mo," which suggest connections to themes of loss, longing, or disappointment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

ties

-0.20

issen

-0.16

MI

-0.15

.cond

-0.15

esan

-0.15

gies

-0.14

hoot

-0.14

opoulos

-0.14

 drafting

-0.14

 construct

-0.14

POSITIVE LOGITS

imum

0.23

moth

0.21

chine

0.20

ifest

0.19

luet

0.18

kest

0.17

ufact

0.17

tering

0.17

ander

0.16

mor

0.16

Activations Density 0.076%

words related to "mar" and "mo," which suggest connections to themes of loss, longing, or disappointment

No Comments

No Known Activations