INDEX

Explanations

instances of the word "it" and variations of "is" or "was" used within various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

_utilities

-0.15

Hang

-0.15

borg

-0.15

Generated

-0.15

generated

-0.14

ØªØ±ÛĮÙĨ

-0.14

 hang

-0.14

 emerg

-0.13

 Tits

-0.13

POSITIVE LOGITS

chy

0.19

oud

0.17

ching

0.16

ÛĮØ§Øª

0.15

olders

0.15

xis

0.15

 happen

0.15

rog

0.14

 happens

0.14

é½

0.14

Activations Density 0.207%

instances of the word "it" and variations of "is" or "was" used within various contexts

No Comments

No Known Activations