INDEX

Explanations

text related to the description and characteristics of fictional characters or entities, potentially from a script or storyline

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

milo

-0.96

ciating

-0.69

dinand

-0.66

 Engineers

-0.63

emouth

-0.61

 pige

-0.61

ONSORED

-0.60

soDeliveryDate

-0.59

 mosqu

-0.59

Interstitial

-0.58

POSITIVE LOGITS

1.20

],

0.91

])

0.89

 edit

0.82

].

0.82

][

0.81

Spoiler

0.75

.)

0.75

itals

0.73

ģ«

0.70

Activations Density 0.857%

text related to the description and characteristics of fictional characters or entities, potentially from a script or storyline

No Comments

No Known Activations