INDEX

Explanations

phrases where someone is instructing or advising someone else

oai_token-act-pair · gpt-3.5-turbo

the preposition "to" in various contexts

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

 divides

-0.67

hops

-0.67

 widened

-0.66

 licenses

-0.65

 accelerated

-0.62

 netted

-0.61

 divisions

-0.61

 Provision

-0.60

 partitions

-0.60

 deductions

-0.59

POSITIVE LOGITS

wered

1.12

pless

0.99

othy

0.97

asted

0.92

asts

0.89

lling

0.89

ilet

0.88

ller

0.87

ffee

0.85

 remind

0.85

Activations Density 0.328%

phrases where someone is instructing or advising someone else

the preposition "to" in various contexts

No Comments

No Known Activations