INDEX

Explanations

phrases related to significant or important information

oai_token-act-pair · gpt-3.5-turbo

phrases indicating important topics and issues

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

 acquies

-0.62

 chalk

-0.61

 pend

-0.60

 meanwhile

-0.59

identally

-0.58

sbm

-0.56

 Lumpur

-0.56

 largeDownload

-0.55

 capit

-0.54

 outnumbered

-0.54

POSITIVE LOGITS

iosyn

0.77

 matter

0.65

 mattered

0.64

 separates

0.62

 surrounds

0.61

 matters

0.60

utra

0.59

oras

0.59

arsity

0.58

orn

0.58

Activations Density 0.029%

phrases related to significant or important information

phrases indicating important topics and issues

No Comments

No Known Activations