INDEX

Explanations

mentions of features or specifications in a description

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L8R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.8.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

urement

-0.15

Ĵáŀ

-0.15

ifiable

-0.14

aight

-0.14

ym

-0.14

izable

-0.14

ares

-0.14

ales

-0.13

ocrine

-0.13

POSITIVE LOGITS

ichen

0.17

eland

0.16

.dup

0.16

áº½

0.14

rench

0.14

urm

0.14

IELD

0.14

Ĥ¬

0.14

odb

0.14

emouth

0.14

Activations Density 0.085%

mentions of features or specifications in a description

No Comments

No Known Activations