INDEX

Explanations

instances of the word "say" or its variations and related expressions indicating speech or assertion

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

ikk

-0.73

summary

-0.72

appers

-0.70

iets

-0.66

oggle

-0.66

-0.65

rows

-0.65

uga

-0.64

onds

-0.63

obar

-0.63

POSITIVE LOGITS

 coinc

0.68

...?

0.59

 therein

0.58

)?

0.58

...)

0.57

ably

0.57

inc

0.56

eth

0.56

âĢ¦)

0.56

Ħ¢

0.56

Activations Density 0.025%

instances of the word "say" or its variations and related expressions indicating speech or assertion

No Comments

No Known Activations