INDEX

Explanations

expressions related to a progression or sequential steps, particularly involving interactions between individuals

oai_token-act-pair · gpt-3.5-turbo

instances of the word "one."

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

folk

-0.74

hips

-0.73

inders

-0.69

ographics

-0.66

 Corpus

-0.66

mson

-0.66

ãĥ©ãĥ³

-0.65

aeda

-0.65

culus

-0.65

esta

-0.64

POSITIVE LOGITS

day

1.15

 minute

1.05

 hundred

1.02

 hour

0.90

 month

0.88

way

0.86

 night

0.85

 thousand

0.85

 inch

0.84

 week

0.83

Activations Density 0.067%

expressions related to a progression or sequential steps, particularly involving interactions between individuals

instances of the word "one."

No Comments

No Known Activations