INDEX

Explanations

words related to microbiology or medical conditions related to pathogens

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ĪĴ

-0.71

ħĭ

-0.64

 Uriel

-0.62

enger

-0.60

 Dish

-0.58

Reviewer

-0.58

 Voyager

-0.58

RTX

-0.58

 Fraz

-0.57

FW

-0.57

POSITIVE LOGITS

entric

1.32

kefeller

1.25

hester

1.18

abulary

1.08

urrencies

1.06

lean

1.05

arbon

1.02

ombat

1.02

ourse

0.98

chio

0.98

Activations Density 0.021%

words related to microbiology or medical conditions related to pathogens

No Comments

No Known Activations