INDEX

Explanations

phrases related to giving instructions or advice using the word "so."

oai_token-act-pair · gpt-3.5-turbo

conjunctions that indicate purpose or result

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

theless

-0.64

 Flavoring

-0.59

esi

-0.58

inch

-0.57

 glances

-0.56

wire

-0.55

wreck

-0.55

question

-0.55

ãĥ¯ãĥ³

-0.54

exclusive

-0.54

POSITIVE LOGITS

othe

1.19

bered

1.17

apy

1.06

othes

1.00

oths

0.92

aps

0.88

arer

0.85

oner

0.85

ãģ®éŃĶ

0.77

pping

0.77

Activations Density 0.077%

phrases related to giving instructions or advice using the word "so."

conjunctions that indicate purpose or result

No Comments

No Known Activations