INDEX

Explanations

text related to quantitative information, measurement, and numerical comparisons

oai_token-act-pair · gpt-3.5-turbo

expressions related to measurement and evaluation criteria

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

rique

-0.77

rek

-0.75

itar

-0.73

 disgr

-0.72

çīĪ

-0.70

 pard

-0.69

 bailed

-0.69

 reborn

-0.69

©¶æ

-0.68

 celebr

-0.67

POSITIVE LOGITS

 determines

1.16

 determine

1.10

 determining

1.02

 measurement

1.02

 calculate

1.00

 varies

0.96

 Calcul

0.93

 estimation

0.88

 subjective

0.87

etermin

0.86

Activations Density 0.940%

text related to quantitative information, measurement, and numerical comparisons

expressions related to measurement and evaluation criteria

No Comments

No Known Activations