INDEX

Explanations

verbs related to actions and intentions in the context of decision-making or progress

oai_token-act-pair · gpt-3.5-turbo

verbs and phrases expressing intentions or desires

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

successfully

-0.76

etheus

-0.71

olor

-0.69

inducing

-0.69

lav

-0.67

egal

-0.67

ulet

-0.66

 notably

-0.65

obook

-0.65

later

-0.64

POSITIVE LOGITS

 darn

0.81

 plain

0.78

ifiable

0.74

 Seah

0.68

cks

0.66

 anecdotal

0.65

ATHER

0.64

 coincidence

0.63

sad

0.63

Sov

0.62

Activations Density 0.375%

verbs related to actions and intentions in the context of decision-making or progress

verbs and phrases expressing intentions or desires

No Comments

No Known Activations