INDEX

Explanations

words related to employment and job situations

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

 Revision

-0.85

 Flavoring

-0.69

Oo

-0.66

 Totem

-0.64

gran

-0.64

theless

-0.63

 sidx

-0.60

LECT

-0.60

 Blast

-0.60

Interstitial

-0.59

POSITIVE LOGITS

 jobs

0.92

hare

0.80

uilt

0.80

paces

0.78

jobs

0.78

aving

0.75

hops

0.75

days

0.74

 unemployed

0.74

aday

0.74

Activations Density 0.050%

words related to employment and job situations

No Comments

No Known Activations