INDEX

Explanations

references to numerical sequences, particularly involving repeated actions or iterations

oai_token-act-pair · gpt-3.5-turbo

terms related to gameplay mechanics and tracking in competitive or turn-based games

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.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

 notor

-0.69

 conspicuous

-0.65

regon

-0.63

Bir

-0.62

cyn

-0.62

 Flavoring

-0.61

 conscientious

-0.61

 indemn

-0.61

 housing

-0.61

 unaff

-0.61

POSITIVE LOGITS

Ago

0.97

clock

0.82

 onwards

0.77

oz

0.76

sequence

0.76

 duration

0.73

 consecut

0.73

 Sequence

0.73

Timeout

0.72

ago

0.71

Activations Density 0.174%

references to numerical sequences, particularly involving repeated actions or iterations

terms related to gameplay mechanics and tracking in competitive or turn-based games

No Comments

No Known Activations