INDEX

Explanations

references to an adversarial context or situation

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

flo

-0.81

regon

-0.81

ajo

-0.79

head

-0.74

uesday

-0.74

otide

-0.71

ikk

-0.70

inker

-0.70

stones

-0.69

psc

-0.69

POSITIVE LOGITS

arial

1.16

 advers

1.12

 adversary

0.90

 adversaries

0.85

 posture

0.77

 posed

0.72

 moderators

0.72

 handshake

0.69

 defences

0.67

 dilig

0.67

Activations Density 0.010%

references to an adversarial context or situation

No Comments

No Known Activations