INDEX

Explanations

phrases related to policy changes or legal decisions that have significant impact or consequences

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

OTOS

-0.72

RAW

-0.71

odore

-0.70

anooga

-0.70

aucuses

-0.69

arel

-0.69

urses

-0.69

icted

-0.68

essions

-0.67

frey

-0.65

POSITIVE LOGITS

 wedge

0.77

 separating

0.77

otomy

0.73

 whereby

0.73

 fallacy

0.73

 mentality

0.73

 Revis

0.72

 medi

0.69

 metast

0.68

pie

0.66

Activations Density 2.344%

phrases related to policy changes or legal decisions that have significant impact or consequences

No Comments

No Known Activations