INDEX

Explanations

terms related to forecasting and predictions

oai_token-act-pair · gpt-3.5-turbo

the word "Fore" and its varying capitalization patterns

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

å°Ĩ

-0.91

naires

-0.88

Els

-0.82

illin

-0.76

REDACTED

-0.74

HAEL

-0.71

IRO

-0.70

Magikarp

-0.69

VG

-0.66

naire

-0.66

POSITIVE LOGITS

sts

1.23

nsics

1.18

closure

1.13

gotten

1.02

runner

1.01

nsic

0.99

gone

0.97

closed

0.97

stal

0.96

shore

0.95

Activations Density 0.014%

terms related to forecasting and predictions

the word "Fore" and its varying capitalization patterns

No Comments

No Known Activations