INDEX

Explanations

prepositions that indicate a point in time

oai_token-act-pair · gpt-3.5-turbo

references to time and conditional phrases relating to events

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

Joy

-0.81

atl

-0.74

atus

-0.72

Reviewer

-0.71

orers

-0.70

atform

-0.69

orer

-0.69

vv

-0.64

 Lawyers

-0.62

lance

-0.61

POSITIVE LOGITS

 alike

0.72

senal

0.63

eele

0.62

 acquaintance

0.62

 sibling

0.62

 transact

0.60

Malley

0.60

 personality

0.60

 relationship

0.60

 dependent

0.58

Activations Density 0.247%

prepositions that indicate a point in time

references to time and conditional phrases relating to events

No Comments

No Known Activations