INDEX

Explanations

references to analysis and comparison involving different categories or entities

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

-0.17

-0.15

æ´²

-0.14

ween

-0.14

.rel

-0.14

rig

-0.14

void

-0.14

sqr

-0.14

 Feinstein

-0.14

pert

-0.13

POSITIVE LOGITS

ÏģÎ¯

0.18

esper

0.15

uchs

0.15

THREAD

0.15

ahoma

0.15

 tá»¥

0.15

erse

0.14

ÏĩÎ½

0.14

phalt

0.14

klu

0.14

Activations Density 0.006%

references to analysis and comparison involving different categories or entities

No Comments

No Known Activations