INDEX

Explanations

terms related to events or actions that follow or result from something else

oai_token-act-pair · gpt-3.5-turbo

the word "subsequent" and its variations in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Haram

-0.85

rod

-0.79

tics

-0.77

eye

-0.76

Bay

-0.75

lua

-0.75

toe

-0.69

mberg

-0.69

park

-0.69

eping

-0.69

POSITIVE LOGITS

 generations

1.04

 iterations

1.00

 installments

0.97

wcs

0.89

 batches

0.89

 unsuccessful

0.86

ĸļ

0.85

 editions

0.85

 revisions

0.84

 subsequent

0.83

Activations Density 0.004%

terms related to events or actions that follow or result from something else

the word "subsequent" and its variations in different contexts

No Comments

No Known Activations