INDEX

Explanations

phrases related to intentions or plans

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate intent or determination to proceed with actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ortment

-0.64

esse

-0.61

 Enhancement

-0.60

legged

-0.59

Kings

-0.58

hazard

-0.58

uli

-0.57

perhaps

-0.57

peria

-0.57

cius

-0.57

POSITIVE LOGITS

 anymore

1.58

 anywhere

1.13

nor

1.12

 bother

1.06

 whatsoever

1.04

 anybody

1.04

 necessarily

1.01

 anything

0.99

 slightest

0.99

any

0.98

Activations Density 0.162%

phrases related to intentions or plans

phrases that indicate intent or determination to proceed with actions

No Comments

No Known Activations