INDEX

Explanations

words related to acceptance

oai_token-act-pair · gpt-3.5-turbo

terms related to acceptance and development processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ULTS

-0.72

 nerv

-0.66

FORE

-0.65

Zam

-0.65

YS

-0.64

hol

-0.62

lap

-0.62

 hour

-0.61

OPLE

-0.60

aimon

-0.60

POSITIVE LOGITS

ed

1.93

edly

1.57

ing

1.42

edIn

1.30

edin

1.13

ships

1.11

ers

1.06

ered

1.05

er

1.03

Activations Density 0.066%

words related to acceptance

terms related to acceptance and development processes

No Comments

No Known Activations