INDEX

Explanations

terms related to challenges and difficulties in 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 @ 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

itz

-0.15

olie

-0.15

erie

-0.15

hr

-0.14

 undert

-0.14

hec

-0.14

_accessible

-0.14

ÑĢÐ°Ð½Ð°

-0.14

mt

-0.14

 ApplicationController

-0.13

POSITIVE LOGITS

 Salisbury

0.16

 futuro

0.16

 Benjamin

0.14

kli

0.14

 Baton

0.14

/pg

0.14

zal

0.14

Lud

0.14

 Anita

0.14

 Spoon

0.13

Activations Density 0.007%

terms related to challenges and difficulties in various contexts

No Comments

No Known Activations