INDEX

Explanations

the word "what" 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 @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

hots

-0.16

elman

-0.15

antly

-0.15

han

-0.14

deen

-0.14

shaw

-0.14

iÃ³n

-0.14

 stiff

-0.13

elt

-0.13

POSITIVE LOGITS

lesh

0.17

IDX

0.17

IDX

0.15

ampoo

0.15

$MESS

0.15

Ð¾ÑĩÐ½Ð¾

0.15

annah

0.14

washing

0.14

NCY

0.14

'gc

0.14

Activations Density 0.083%

the word "what" in various contexts

No Comments

No Known Activations