INDEX

Explanations

phrases starting with "And."

oai_token-act-pair · gpt-3.5-turbo

instances of the word "And" used in connection to various statements or clauses

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

manship

-0.66

ded

-0.66

äºº

-0.65

BILITIES

-0.65

scene

-0.65

-0.64

atical

-0.64

[&

-0.64

houses

-0.63

-0.62

POSITIVE LOGITS

romeda

1.39

hra

1.08

rea

1.08

alus

0.91

rew

0.87

 secondly

0.85

raltar

0.81

 then

0.80

alys

0.80

ersen

0.79

Activations Density 0.050%

phrases starting with "And."

instances of the word "And" used in connection to various statements or clauses

No Comments

No Known Activations