INDEX

Explanations

numerical references related to specific time frames or durations

oai_token-act-pair · gpt-3.5-turbo

phrases indicating sequences of time or periods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

tics

-0.76

ucl

-0.68

ingen

-0.66

âĶģ

-0.64

pherd

-0.64

borgh

-0.61

mbuds

-0.59

borg

-0.58

morph

-0.58

à©

-0.58

POSITIVE LOGITS

 half

1.12

few

1.12

 couple

1.02

 quarter

1.02

 installment

0.97

 responders

0.95

 phase

0.93

half

0.93

decade

0.92

two

0.92

Activations Density 0.081%

numerical references related to specific time frames or durations

phrases indicating sequences of time or periods

No Comments

No Known Activations