INDEX

Explanations

names of specific locations, possibly related to politics or events happening in those locations

oai_token-act-pair · gpt-3.5-turbo

references to the place or entity "Samar"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

RAFT

-1.04

ered

-0.82

spring

-0.80

ORN

-0.77

sheet

-0.73

puff

-0.72

draft

-0.71

cember

-0.70

urden

-0.70

ORED

-0.70

POSITIVE LOGITS

itans

1.09

itan

1.03

 Samar

0.93

pling

0.84

eness

0.78

agi

0.76

pha

0.75

elines

0.72

autions

0.72

anthrop

0.71

Activations Density 0.009%

names of specific locations, possibly related to politics or events happening in those locations

references to the place or entity "Samar"

No Comments

No Known Activations