INDEX

Explanations

political and governmental references

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

LD

-0.66

MPG

-0.65

rists

-0.59

Hog

-0.58

LX

-0.58

-0.57

geons

-0.57

 veter

-0.56

 multic

-0.56

patient

-0.56

POSITIVE LOGITS

window

0.75

farious

0.69

Speaking

0.66

ãĤ¨ãĥ«

0.65

ensation

0.64

Asked

0.63

ado

0.63

]).

0.63

ocaly

0.62

uchin

0.62

Activations Density 1.379%

political and governmental references

No Comments

No Known Activations