INDEX

Explanations

first-person pronouns and references to personal experiences

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

apl

-0.16

ihan

-0.15

ocode

-0.15

VENTORY

-0.14

Č↵

-0.14

 ÄĮeskÃ¡

-0.14

ools

-0.13

nbsp

-0.13

estado

-0.13

áºŃy

-0.13

POSITIVE LOGITS

 pods

0.15

vb

0.15

ãģĹãĤĩ

0.14

riott

0.14

ìĦł

0.14

 besides

0.14

ipsis

0.13

 seeds

0.13

OUSE

0.13

 germ

0.13

Activations Density 0.450%

first-person pronouns and references to personal experiences

No Comments

No Known Activations