INDEX

Explanations

proper nouns, such as names of individuals or specific entities

oai_token-act-pair · gpt-3.5-turbo

unique identifiers or specific terms related to various subjects

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

],"

-0.76

interstitial

-0.76

anmar

-0.64

ibaba

-0.62

azard

-0.58

CFL

-0.56

anchester

-0.56

 epid

-0.55

',"

-0.55

veyard

-0.54

POSITIVE LOGITS

 meanwhile

0.80

 however

0.79

 ---------

0.74

udos

0.74

 solves

0.71

beit

0.71

 refers

0.70

 assures

0.70

 disagrees

0.67

 emphasizes

0.67

Activations Density 0.677%

proper nouns, such as names of individuals or specific entities

unique identifiers or specific terms related to various subjects

No Comments

No Known Activations