INDEX

Explanations

abbreviations related to technology and data processing

oai_token-act-pair · gpt-3.5-turbo

references to "DP," likely indicating specific data points or metrics

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

glers

-0.84

taboola

-0.78

ton

-0.76

ãĤ¤ãĥĪ

-0.74

umn

-0.74

swick

-0.73

bee

-0.73

itud

-0.70

ãĥĥ

-0.70

combe

-0.69

POSITIVE LOGITS

ressed

1.02

rint

0.98

olicy

0.97

onent

0.88

odcast

0.87

ORTS

0.86

arent

0.85

ablo

0.82

orter

0.81

ressing

0.81

Activations Density 0.006%

abbreviations related to technology and data processing

references to "DP," likely indicating specific data points or metrics

No Comments

No Known Activations