INDEX

Explanations

mentions of the word "pod"

oai_token-act-pair · gpt-3.5-turbo

references to "pod" or related terms within the text

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

ãĤ¤ãĥĪ

-0.72

FUL

-0.72

ãĤ¼ãĤ¦ãĤ¹

-0.67

listed

-0.67

ãĤ´ãĥ³

-0.66

å°Ĩ

-0.63

 hurdles

-0.62

 defic

-0.62

ãĥ¼ãĥ³

-0.61

 WARN

-0.60

POSITIVE LOGITS

esta

1.07

casters

1.06

iatric

1.06

Pod

0.97

caster

0.96

ols

0.93

pod

0.91

ophy

0.86

ikarp

0.86

omatic

0.85

Activations Density 0.013%

mentions of the word "pod"

references to "pod" or related terms within the text

No Comments

No Known Activations