INDEX

Explanations

phrases that indicate a relationship to a "based" concept or idea

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

shire

-0.79

OUGH

-0.76

ubb

-0.74

IVER

-0.74

osure

-0.73

olit

-0.73

imbabwe

-0.73

osen

-0.72

coni

-0.72

erved

-0.72

POSITIVE LOGITS

 therapies

0.97

 solutions

0.91

 economy

0.87

 systems

0.86

 therapy

0.85

 crowdfunding

0.84

 harassment

0.83

 approach

0.83

 strategy

0.83

 interrogation

0.83

Activations Density 0.021%

phrases that indicate a relationship to a "based" concept or idea

No Comments

No Known Activations