INDEX

Explanations

action-related verbs

oai_token-act-pair · gpt-3.5-turbo

narrative shifts or changes in a storyline

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

fter

-0.65

guide

-0.64

dor

-0.60

arta

-0.59

FIELD

-0.59

pta

-0.58

photo

-0.58

toggle

-0.58

 IMAGES

-0.57

orio

-0.57

POSITIVE LOGITS

 raining

1.03

 uphill

0.68

 downhill

0.66

 impossible

0.64

 easier

0.61

 incre

0.59

 spur

0.55

 worthwhile

0.54

 advisable

0.54

 corro

0.54

Activations Density 1.110%

action-related verbs

narrative shifts or changes in a storyline

No Comments

No Known Activations