INDEX

Explanations

demonstrative pronouns and phrases indicating specific references or clarifications

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

terday

-0.80

å§«

-0.78

ilaterally

-0.75

Ń·

-0.73

cycles

-0.70

sample

-0.67

ctors

-0.66

Īè

-0.66

options

-0.66

geons

-0.65

POSITIVE LOGITS

 latter

0.76

 vein

0.75

 same

0.74

 pecul

0.70

 type

0.69

 sort

0.68

 visceral

0.66

 simple

0.66

 perverse

0.66

tru

0.66

Activations Density 0.025%

demonstrative pronouns and phrases indicating specific references or clarifications

No Comments

No Known Activations