INDEX

Explanations

pronouns and references to personal involvement or relationships

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

byt

-0.15

uae

-0.14

mask

-0.13

_locals

-0.13

uin

-0.13

 Marino

-0.13

cop

-0.13

ligt

-0.13

vero

-0.13

 sleeves

-0.13

POSITIVE LOGITS

oti

0.17

redential

0.16

lox

0.15

CLU

0.15

ARING

0.14

ÅĽcie

0.14

Ð²Ð°ÑĢ

0.14

 Abed

0.14

obuf

0.14

Activations Density 0.142%

pronouns and references to personal involvement or relationships

No Comments

No Known Activations