INDEX

Explanations

names related to investigations, particularly focusing on the figure of Robert Mueller and his role in political scandals

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

Pyr

-0.87

 Hawaiian

-0.70

 Zoro

-0.67

 Columb

-0.67

 Fiji

-0.66

piring

-0.65

UX

-0.65

 Celtic

-0.65

 Barbarian

-0.65

 Esper

-0.65

POSITIVE LOGITS

 Mueller

1.02

 Probe

0.91

ueller

0.90

 investigation

0.88

angelo

0.86

 probe

0.85

 investigations

0.84

owitz

0.83

 indict

0.81

 investigating

0.80

Activations Density 0.005%

names related to investigations, particularly focusing on the figure of Robert Mueller and his role in political scandals

No Comments

No Known Activations