INDEX

Explanations

occurrences of formal assertions or statements about events and their 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

 ILCS

-0.76

ategory

-0.70

ordinate

-0.67

everal

-0.66

 hemor

-0.63

ccording

-0.62

 veter

-0.62

minster

-0.61

 administ

-0.61

 belonging

-0.59

POSITIVE LOGITS

 echoed

0.81

 illustrates

0.79

 ignores

0.76

 coincides

0.74

 succeeds

0.74

 differs

0.72

 spree

0.72

 contradicts

0.72

oids

0.71

 underscores

0.70

Activations Density 0.136%

occurrences of formal assertions or statements about events and their contexts

No Comments

No Known Activations