INDEX

Explanations

specific phrases starting with "In a" followed by an adjective or noun

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "In a" followed by various contextual descriptors

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 incompet

-0.65

 quotas

-0.64

 aggress

-0.64

angering

-0.63

auder

-0.63

 excellence

-0.63

 recruits

-0.62

 icons

-0.62

 favourites

-0.62

 offerings

-0.62

POSITIVE LOGITS

 nutshell

1.57

 sense

0.95

 pinch

0.85

uckland

0.83

 perverse

0.81

 twist

0.80

 sane

0.80

 typical

0.79

 vacuum

0.79

 society

0.77

Activations Density 0.080%

specific phrases starting with "In a" followed by an adjective or noun

instances of the phrase "In a" followed by various contextual descriptors

No Comments

No Known Activations