INDEX

Explanations

information related to statistics or data analysis

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

sort

-0.69

alph

-0.66

andal

-0.64

pedia

-0.64

spir

-0.63

hunt

-0.63

Cru

-0.62

abl

-0.61

gang

-0.59

hement

-0.59

POSITIVE LOGITS

GDP

1.05

 respondents

0.98

 total

0.93

 households

0.87

 variance

0.87

 earners

0.84

 Americans

0.82

 surveyed

0.80

 population

0.79

 overall

0.77

Activations Density 0.579%

information related to statistics or data analysis

No Comments

No Known Activations