INDEX

Explanations

references to specific movies and characters within those movies, as well as information related to the film industry

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

taboola

-0.83

 guiActiveUn

-0.78

Tos

-0.70

 Expend

-0.67

BILITIES

-0.66

MX

-0.66

Cola

-0.64

ħĭ

-0.62

©¶æ

-0.61

 Resp

-0.61

POSITIVE LOGITS

inki

0.91

schild

0.77

igans

0.75

warts

0.75

quarters

0.73

sung

0.69

aunts

0.69

iday

0.67

awei

0.66

 Salam

0.65

Activations Density 2.801%

references to specific movies and characters within those movies, as well as information related to the film industry

No Comments

No Known Activations