INDEX

Explanations

instances of the word "we" and its variations, indicating a focus on collective statements or reflections

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

qd

-0.15

.RunWith

-0.15

adies

-0.14

apy

-0.14

aucoup

-0.14

opian

-0.14

 speculate

-0.14

ãĥ¼ãĥĨ

-0.14

antro

-0.14

ylko

-0.13

POSITIVE LOGITS

 must

0.19

 shall

0.18

can

0.17

 might

0.17

know

0.17

 know

0.16

 Talking

0.16

ÑĤÑĮ

0.16

 talked

0.16

 talking

0.15

Activations Density 0.098%

instances of the word "we" and its variations, indicating a focus on collective statements or reflections

No Comments

No Known Activations