INDEX

Explanations

words related to actions or processes, particularly involving physical activities or operations

oai_token-act-pair · gpt-3.5-turbo

gerunds and participles related to action and process

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

QB

-0.66

pps

-0.65

fitting

-0.63

spot

-0.60

bal

-0.60

bold

-0.60

trak

-0.59

ulet

-0.59

ngth

-0.58

Baz

-0.57

POSITIVE LOGITS

redients

1.20

tons

1.08

HAM

0.97

ame

0.75

 spree

0.73

 aids

0.72

ham

0.72

 noises

0.71

roup

0.69

quished

0.69

Activations Density 0.172%

words related to actions or processes, particularly involving physical activities or operations

gerunds and participles related to action and process

No Comments

No Known Activations