INDEX

Explanations

phrases indicating a predominant or general trend or state

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

aise

-0.16

asse

-0.16

-0.15

_UNICODE

-0.15

_traits

-0.14

enia

-0.14

ires

-0.14

 sometimes

-0.14

alted

-0.14

uong

-0.14

POSITIVE LOGITS

stay

0.15

à¹Ĩ

0.15

MMdd

0.14

 èī¯

0.14

ĽĦ

0.14

overlap

0.14

è¿ĺæĺ¯

0.14

/md

0.14

LY

0.14

 Mostly

0.14

Activations Density 0.041%

phrases indicating a predominant or general trend or state

No Comments

No Known Activations