INDEX

Explanations

words related to describing how something is done or carried out

oai_token-act-pair · gpt-3.5-turbo

forms of the verb "to be" in various tenses

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

 Clarks

-0.75

osate

-0.71

aldehyde

-0.68

çīĪ

-0.66

 Rouge

-0.63

eware

-0.62

 Davidson

-0.62

teness

-0.60

Chan

-0.60

cpp

-0.59

POSITIVE LOGITS

 fared

1.15

 handled

1.04

 progressing

0.95

 structured

0.95

 impacted

0.91

 unfolded

0.90

 shaping

0.89

 behaved

0.89

 interact

0.88

 affect

0.87

Activations Density 0.157%

words related to describing how something is done or carried out

forms of the verb "to be" in various tenses

No Comments

No Known Activations