INDEX

Explanations

keywords related to specificity or targeting

oai_token-act-pair · gpt-3.5-turbo

phrases or words that indicate specificity or detailed descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

OWER

-0.70

NER

-0.70

 Feldman

-0.67

 Dinosaur

-0.65

=-=-=-=-=-=-=-=-

-0.65

rican

-0.64

ORN

-0.64

 Alive

-0.64

MER

-0.63

 Springer

-0.63

POSITIVE LOGITS

ivity

1.00

ities

0.97

ally

0.96

iveness

0.94

arily

0.93

ivities

0.89

ively

0.86

iations

0.85

ality

0.84

ancial

0.84

Activations Density 0.007%

keywords related to specificity or targeting

phrases or words that indicate specificity or detailed descriptions

No Comments

No Known Activations