INDEX

Explanations

references to an organization or group, particularly those denoted by "AG" followed by numbers in various contexts

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

lihood

-0.82

 sight

-0.69

iculty

-0.67

phrine

-0.63

tons

-0.62

 Borough

-0.62

around

-0.60

 Seas

-0.60

dates

-0.59

 Gorge

-0.58

POSITIVE LOGITS

reement

1.08

ENCY

1.08

REE

0.99

reements

0.92

ENC

0.92

EMENT

0.90

reed

0.88

VERTISEMENT

0.84

HHHH

0.83

oras

0.82

Activations Density 0.003%

references to an organization or group, particularly those denoted by "AG" followed by numbers in various contexts

No Comments

No Known Activations