INDEX

Explanations

statements of fact or claims regarding assessments or evaluations

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

 welf

-0.65

umbnail

-0.62

 Telesc

-0.62

umbnails

-0.61

ukong

-0.61

});

-0.61

Owl

-0.60

 sang

-0.60

 Compan

-0.60

 traveler

-0.59

POSITIVE LOGITS

 unclear

1.17

 impossible

1.12

 unlikely

1.05

 possible

1.04

 conceivable

0.96

 imperative

0.94

 difficult

0.94

 advisable

0.91

 feasible

0.90

 unsur

0.90

Activations Density 0.079%

statements of fact or claims regarding assessments or evaluations

No Comments

No Known Activations