INDEX

Explanations

references to medical conditions and their implications

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

 Franch

-0.77

Catalog

-0.69

Theme

-0.69

 Platform

-0.68

Flags

-0.67

Untitled

-0.66

selage

-0.66

Lenin

-0.66

Platform

-0.65

 Pirates

-0.65

POSITIVE LOGITS

 symptoms

1.25

 hormonal

1.18

 clinically

1.17

 Symptoms

1.17

 hormone

1.16

 patients

1.14

 diabetic

1.13

 lesions

1.11

 medication

1.10

 physiological

1.09

Activations Density 0.655%

references to medical conditions and their implications

No Comments

No Known Activations