INDEX

Explanations

short phrases starting with "It" and describing an item or experience in qualitative terms

oai_token-act-pair · gpt-3.5-turbo

instances of the word "It" used in positive contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

anton

-0.68

Ĥ¬

-0.63

 Corpus

-0.61

dding

-0.60

 dstg

-0.59

 persuasion

-0.58

Priv

-0.57

å½

-0.57

 ILCS

-0.56

hips

-0.56

POSITIVE LOGITS

 contains

1.20

alian

1.15

 seems

1.14

's

1.09

 consists

1.09

unes

1.08

'll

1.02

 wasn

1.01

 incorporates

1.01

 resembles

1.01

Activations Density 0.295%

short phrases starting with "It" and describing an item or experience in qualitative terms

instances of the word "It" used in positive contexts

No Comments

No Known Activations