INDEX

Explanations

references to legal or legislative matters

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

lections

-0.65

cheat

-0.65

 Reincarnated

-0.63

boro

-0.60

apon

-0.60

heit

-0.60

tons

-0.59

geries

-0.58

 endeav

-0.58

naissance

-0.58

POSITIVE LOGITS

 Related

0.87

ÂłÂłÂłÂł

0.76

ÂłÂłÂł

0.76

Meanwhile

0.72

Speaking

0.71

Asked

0.71

Correct

0.67

Indeed

0.67

However

0.66

Earlier

0.66

Activations Density 0.096%

references to legal or legislative matters

No Comments

No Known Activations