INDEX

Explanations

words related to events or actions with a negative connotation or conflict

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Ô

-0.87

quickShipAvailable

-0.79

iband

-0.79

enta

-0.79

brace

-0.74

ãĥ´

-0.74

ãģ¯

-0.73

knit

-0.72

ipedia

-0.71

say

-0.71

POSITIVE LOGITS

 spree

1.05

 extraord

1.00

 prowess

0.88

 antics

0.88

 frenzy

0.87

 activities

0.86

 behaviour

0.86

 activity

0.84

 habits

0.83

 abilities

0.83

Activations Density 0.104%

words related to events or actions with a negative connotation or conflict

No Comments

No Known Activations