INDEX

Explanations

adjectives describing the quality or state of something

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

HAEL

-0.89

venant

-0.79

adium

-0.75

 DRAG

-0.73

PLIED

-0.73

IAS

-0.71

elsen

-0.70

upon

-0.69

Journal

-0.69

RY

-0.69

POSITIVE LOGITS

 nifty

1.03

 darn

1.01

nice

0.99

 neat

0.92

 tame

0.91

 cool

0.88

 alright

0.88

 straightforward

0.87

 harmless

0.87

 snipp

0.86

Activations Density 0.746%

adjectives describing the quality or state of something

No Comments

No Known Activations