INDEX

Explanations

references to people and pronouns, indicating a focus on interpersonal relationships and perspectives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

heimer

-0.17

vey

-0.16

RLF

-0.15

 gotta

-0.14

 musel

-0.14

Must

-0.14

must

-0.14

å¿ħé¡»

-0.14

anter

-0.14

cannot

-0.14

POSITIVE LOGITS

 should

0.50

sh

0.48

should

0.44

 Should

0.43

SH

0.43

SH

0.42

sho

0.42

.should

0.40

Should

0.39

_should

0.39

Activations Density 0.095%

references to people and pronouns, indicating a focus on interpersonal relationships and perspectives

No Comments

No Known Activations