INDEX

Explanations

articles and quantifiers indicating significance or recentness in descriptions

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

ilde

-0.16

ht

-0.15

pora

-0.14

_STS

-0.14

 wakes

-0.14

/*č↵

-0.14

 McCart

-0.14

ILA

-0.13

owo

-0.13

uyáº¿n

-0.13

POSITIVE LOGITS

 recent

0.21

 society

0.20

 twist

0.19

 sense

0.18

recent

0.18

 earlier

0.17

 economy

0.16

 nutshell

0.16

nut

0.16

 previous

0.15

Activations Density 0.040%

articles and quantifiers indicating significance or recentness in descriptions

No Comments

No Known Activations